From dcherubangel at gmail.com  Sun Jan  6 08:55:29 2013
From: dcherubangel at gmail.com (Nutthamon)
Date: Sun, 6 Jan 2013 23:55:29 +0700
Subject: [SRILM User List] Do you have character level for LM in SRILM
	toolkit?
Message-ID: <CANFpwQ1Vf64s9jrLoOew09SepF_ESWm2H9rdt_fsq72+yJNG=w@mail.gmail.com>

Hello,

I am new to language modeling and SRILM toolkit.

Is this toolkit can generate language model in character level? If can
do that, what is a command for do that i can't find it.And please give
example to me.

Many thank in advance


-- 
Best Regards,
Nutthamon Moknarong
dcherubangel at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130106/506a69ab/attachment.html>

From dcherubangel at gmail.com  Sun Jan  6 08:56:41 2013
From: dcherubangel at gmail.com (Nutthamon)
Date: Sun, 6 Jan 2013 23:56:41 +0700
Subject: [SRILM User List] Do you have character level for LM in SRILM
	toolkit
Message-ID: <CANFpwQ2tVa=Wc206=zKZBrTcH4xdqSp4EbA_DXQnooBgv+or8g@mail.gmail.com>

Hello,

I am new to language modeling and SRILM toolkit.

Is this toolkit can generate language model in character level? If can
do that, what is a command for do that i can't find it.And please give
example to me.

Many thank in advance


--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130106/eae9df24/attachment.html>

From venkataraman.anand at gmail.com  Sun Jan  6 09:53:17 2013
From: venkataraman.anand at gmail.com (Anand Venkataraman)
Date: Sun, 6 Jan 2013 09:53:17 -0800
Subject: [SRILM User List] Do you have character level for LM in SRILM
	toolkit
In-Reply-To: <CANFpwQ2tVa=Wc206=zKZBrTcH4xdqSp4EbA_DXQnooBgv+or8g@mail.gmail.com>
References: <CANFpwQ2tVa=Wc206=zKZBrTcH4xdqSp4EbA_DXQnooBgv+or8g@mail.gmail.com>
Message-ID: <CAF6FMTXjNwSMnJbRoV=825fRacTAwRr13N1mNQO27Yqk90N0GA@mail.gmail.com>

If you want an LM built over character sequences, you simply have to break
your input stream into whitespace separated letters. Note that there are
many nuances here - E.g. Should you have new word boundary characters
(analogous to <s> and </s>) or simply have one word per line, have the
fixed vocabulary (the alphabet + any meta chars you want) given up front or
learned at build-time, how you handle special characters and punctuations,
etc.

Assuming English text, the following Unix command can get you started. The tr
command breaks the stream into one word per line and sed inserts a space
after every letter on each line.

cat corpus.txt | tr ' ' '\012' | sed 's/\(.\)/\1 /g' | ngram-count ...


HTH

&

On Sun, Jan 6, 2013 at 8:56 AM, Nutthamon <dcherubangel at gmail.com> wrote:

> Hello,
>
> I am new to language modeling and SRILM toolkit.
>
> Is this toolkit can generate language model in character level? If can do that, what is a command for do that i can't find it.And please give example to me.
>
> Many thank in advance
>
>
> --
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130106/4d638427/attachment.html>

From dcherubangel at gmail.com  Mon Jan  7 21:42:18 2013
From: dcherubangel at gmail.com (Koonnoo)
Date: Tue, 8 Jan 2013 12:42:18 +0700
Subject: [SRILM User List] Do you have character level for LM in SRILM
	toolkit
In-Reply-To: <CAF6FMTXjNwSMnJbRoV=825fRacTAwRr13N1mNQO27Yqk90N0GA@mail.gmail.com>
References: <CANFpwQ2tVa=Wc206=zKZBrTcH4xdqSp4EbA_DXQnooBgv+or8g@mail.gmail.com>
	<CAF6FMTXjNwSMnJbRoV=825fRacTAwRr13N1mNQO27Yqk90N0GA@mail.gmail.com>
Message-ID: <187B3C49-1042-4AA2-AED4-C32DC5DCED88@gmail.com>

> Dear Anand
> i used this tool via cygwin terminal.
> 
> Example in training.txt
> 
> s i m p l y
> g o o d
> t h a n k y o u
> c l o u n d
> 
> or
> training2.txt
> s i m p l y g o o d t h a n k y o u c l o u n d
> 
> 
> which training text correct for LM built on character level? first,right? If first i can directly enter to add more line or add some symbol for add line?
> 
> i'm not sure what is <s>and</s> mean.
> 
> Is this command for train lm model character level (trigram)?
> $ ngram-count -text /srilm/training.txt -order 3 -lm /srilm/training.lm
> 
> My english is weak maybe i ask you more than 1 time :)
> thank you in advance
> 
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130108/9003e8cf/attachment.html>

From martaruizcostajussa at gmail.com  Tue Jan  8 03:15:59 2013
From: martaruizcostajussa at gmail.com (Marta Ruiz)
Date: Tue, 8 Jan 2013 19:15:59 +0800
Subject: [SRILM User List] linear interpolation of different vocabulary
	language models
Message-ID: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>

Dear all,

How can I interpolate language models built on the same text but with
different vocabularies. I mean, I have a text with words, lemmas and PoS,
how can I interpolate the language models.

Kind regards,
Marta
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130108/f265a3a5/attachment.html>

From stolcke at icsi.berkeley.edu  Tue Jan  8 12:15:57 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Tue, 08 Jan 2013 12:15:57 -0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
Message-ID: <50EC7E7D.6050703@icsi.berkeley.edu>

On 1/8/2013 3:15 AM, Marta Ruiz wrote:
> Dear all,
>
> How can I interpolate language models built on the same text but with 
> different vocabularies. I mean, I have a text with words, lemmas and PoS,
> how can I interpolate the language models.
You cannot interpolate models that use different types of vocabularies. 
(You could interpolate models that are all word-based but where there 
are differences in the sets of words occurring in the component models.  
The words that are not occurring in some submodel would implicitly have 
probability zero in that submodel).

So what you need to do is:

1. Create a word-based version of each model.  For example, you can 
construct a POS-based LM and combine it with a class membership mapping 
(in classes-format, see man page) to get a word-level POS-based model.   
Similar with lemma-based LMs (the lemmas are effectively word classes).

2. Then interpolate the models using

     ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 .... -lambda ... 
-mix-lambda2 ... -classes CLASSES

where CLASSES is a classes-format(5) file defining the union of all the 
word classes used in the various component models.

Andreas


From martaruizcostajussa at gmail.com  Tue Jan  8 18:07:59 2013
From: martaruizcostajussa at gmail.com (Marta Ruiz)
Date: Wed, 9 Jan 2013 10:07:59 +0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <50EC7E7D.6050703@icsi.berkeley.edu>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
Message-ID: <CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>

Thanks Andreas, two more questions

>
> 1. Create a word-based version of each model.  For example, you can
> construct a POS-based LM and combine it with a class membership mapping (in
> classes-format, see man page) to get a word-level POS-based model.
> Similar with lemma-based LMs (the lemmas are effectively word classes).
>
>
which is the instruction to do this?


> 2. Then interpolate the models using
>
>     ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 .... -lambda ...
> -mix-lambda2 ... -classes CLASSES
>
> where CLASSES is a classes-format(5) file defining the union of all the
> word classes used in the various component models.
>
>
to find the lambdas can I use the compute-best-mix, can't I?

best regards,
Marta



> Andreas
>
>


-- 
Marta Ruiz Costa-juss?
martaruizcostajussa at gmail.com
http://gps-tsc.upc.es/veu/personal/mruiz/mruiz.php3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130109/5079536a/attachment.html>

From dcherubangel at gmail.com  Tue Jan  8 20:19:03 2013
From: dcherubangel at gmail.com (Koonnoo)
Date: Wed, 9 Jan 2013 11:19:03 +0700
Subject: [SRILM User List] Train lm character level
Message-ID: <5BAE282D-2154-437C-BCED-CAFB935E43F4@gmail.com>

> Dear All
> i used this tool via cygwin terminal.
> 
> Example in training.txt
> 
> s i m p l y
> g o o d
> t h a n k y o u
> c l o u n d
> 
> or
> training2.txt
> s i m p l y g o o d t h a n k y o u c l o u n d
> 
> 
> which training text correct for LM built on character level? first,right? If first i can directly enter to add more line or add some symbol for add line?
> 
> i'm not sure what is <s>and</s> mean.
> 
> Is this command for train lm model character level (trigram)?
> $ ngram-count -text /srilm/training.txt -order 3 -lm /srilm/training.lm
> 
> My english is weak maybe i ask you more than 1 time :)
> thank you in advance
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130109/e20fcdce/attachment.html>

From stolcke at icsi.berkeley.edu  Tue Jan  8 23:34:39 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Tue, 08 Jan 2013 23:34:39 -0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
	<CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
Message-ID: <50ED1D8F.7000209@icsi.berkeley.edu>

On 1/8/2013 6:07 PM, Marta Ruiz wrote:
> Thanks Andreas, two more questions
>
>
>     1. Create a word-based version of each model.  For example, you
>     can construct a POS-based LM and combine it with a class
>     membership mapping (in classes-format, see man page) to get a
>     word-level POS-based model.   Similar with lemma-based LMs (the
>     lemmas are effectively word classes).
>
>
> which is the instruction to do this?

1. You create the class-to-word mapping file (in the format described 
here 
<http://www.speech.sri.com/projects/srilm/manpages/classes-format.5.html>) 
to reflect either your POS-to-word or lemma-to-word mapping.
2. Process the training data to replace the words with POS or lemmas, as 
appropriate.
3. Train the ngram portion of the LM by running ngram-count on the 
training data represented as a sequence of POS tags / lemmas (from step 2).


>     2. Then interpolate the models using
>
>         ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 .... -lambda
>     ... -mix-lambda2 ... -classes CLASSES
>
>     where CLASSES is a classes-format(5) file defining the union of
>     all the word classes used in the various component models.
>
>
> to find the lambdas can I use the compute-best-mix, can't I?
Exactly.

Andreas

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130108/2e152197/attachment.html>

From venkataraman.anand at gmail.com  Wed Jan  9 10:25:55 2013
From: venkataraman.anand at gmail.com (Anand Venkataraman)
Date: Wed, 9 Jan 2013 10:25:55 -0800
Subject: [SRILM User List] Train lm character level
In-Reply-To: <5BAE282D-2154-437C-BCED-CAFB935E43F4@gmail.com>
References: <5BAE282D-2154-437C-BCED-CAFB935E43F4@gmail.com>
Message-ID: <CAF6FMTWPfg5_QQjSaTqPU3R99fHDFnTE-LucogCEbwZsj8s7BQ@mail.gmail.com>

It depends on what you want to accomplish with the LM. Under most
circumstances you would want to preserve the word boundary information
(akin to the sentence boundary tags - <s> and </s>, which stands for the
start and  end of a sentence).

The first format you describe (training.txt) accomplishes this by using <s>
and </s> to proxy for your word boundaries. But it loses information which
you might otherwise have obtained from knowledge of which words are likely
to occupy which sentential slots (e.g. the is almost invariably followed by
another word and hence <w> should be more likely after "the" than after,
say, an arbitrary noun.) You could introduce <w> and </w> as special tokens
in training2.txt, for instance.

&

On Tue, Jan 8, 2013 at 8:19 PM, Koonnoo <dcherubangel at gmail.com> wrote:

> Dear All
> i used this tool via cygwin terminal.
>
> Example in training.txt
>
> s i m p l y
> g o o d
> t h a n k y o u
> c l o u n d
>
> or
> training2.txt
> s i m p l y g o o d t h a n k y o u c l o u n d
>
>
> which training text correct for LM built on character level? first,right?
> If first i can directly enter to add more line or add some symbol for add
> line?
>
> i'm not sure what is <s>and</s> mean.
>
> Is this command for train lm model character level (trigram)?
> $ ngram-count -text /srilm/training.txt -order 3 -lm /srilm/training.lm
>
> My english is weak maybe i ask you more than 1 time :)
> thank you in advance
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130109/9d949691/attachment.html>

From martaruizcostajussa at gmail.com  Wed Jan 16 18:00:12 2013
From: martaruizcostajussa at gmail.com (Marta Ruiz)
Date: Thu, 17 Jan 2013 10:00:12 +0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <50ED1D8F.7000209@icsi.berkeley.edu>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
	<CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
	<50ED1D8F.7000209@icsi.berkeley.edu>
Message-ID: <CABEBqH+-y+3CO43oqtqGUFzdbUvB1BtOqLFg1aW964--4JCK=w@mail.gmail.com>

Hi Andreas,

regarding this issue, I got the error

 class definition has too many fields

in fact, I wanted to expand a language model of PoS tags into words...
actually, each PoS has many words related...


best regards,
Marta

On Wed, Jan 9, 2013 at 3:34 PM, Andreas Stolcke
<stolcke at icsi.berkeley.edu>wrote:

>  On 1/8/2013 6:07 PM, Marta Ruiz wrote:
>
> Thanks Andreas, two more questions
>
>>
>> 1. Create a word-based version of each model.  For example, you can
>> construct a POS-based LM and combine it with a class membership mapping (in
>> classes-format, see man page) to get a word-level POS-based model.
>> Similar with lemma-based LMs (the lemmas are effectively word classes).
>>
>>
> which is the instruction to do this?
>
>
> 1. You create the class-to-word mapping file (in the format described here<http://www.speech.sri.com/projects/srilm/manpages/classes-format.5.html>)
> to reflect either your POS-to-word or lemma-to-word mapping.
> 2. Process the training data to replace the words with POS or lemmas, as
> appropriate.
> 3. Train the ngram portion of the LM by running ngram-count on the
> training data represented as a sequence of POS tags / lemmas (from step 2).
>
>
>
>
>
>> 2. Then interpolate the models using
>>
>>     ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 .... -lambda ...
>> -mix-lambda2 ... -classes CLASSES
>>
>> where CLASSES is a classes-format(5) file defining the union of all the
>> word classes used in the various component models.
>>
>>
> to find the lambdas can I use the compute-best-mix, can't I?
>
> Exactly.
>
> Andreas
>
>


-- 
Marta Ruiz Costa-juss?
martaruizcostajussa at gmail.com
http://gps-tsc.upc.es/veu/personal/mruiz/mruiz.php3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130117/5190370a/attachment.html>

From stolcke at icsi.berkeley.edu  Wed Jan 16 19:22:20 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Wed, 16 Jan 2013 19:22:20 -0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <CABEBqH+-y+3CO43oqtqGUFzdbUvB1BtOqLFg1aW964--4JCK=w@mail.gmail.com>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
	<CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
	<50ED1D8F.7000209@icsi.berkeley.edu>
	<CABEBqH+-y+3CO43oqtqGUFzdbUvB1BtOqLFg1aW964--4JCK=w@mail.gmail.com>
Message-ID: <50F76E6C.3020501@icsi.berkeley.edu>

On 1/16/2013 6:00 PM, Marta Ruiz wrote:
> Hi Andreas,
>
> regarding this issue, I got the error
>
>  class definition has too many fields
That means you must have a very long line in your class definitions file.
You should have one class membership definition per line.
If a class has many members you write one per line, for example

NN    cat
NN    dog
NN    ball

etc.

Andreas
>
> in fact, I wanted to expand a language model of PoS tags into words...
> actually, each PoS has many words related...
>
>
> best regards,
> Marta
>
> On Wed, Jan 9, 2013 at 3:34 PM, Andreas Stolcke 
> <stolcke at icsi.berkeley.edu <mailto:stolcke at icsi.berkeley.edu>> wrote:
>
>     On 1/8/2013 6:07 PM, Marta Ruiz wrote:
>>     Thanks Andreas, two more questions
>>
>>
>>         1. Create a word-based version of each model.  For example,
>>         you can construct a POS-based LM and combine it with a class
>>         membership mapping (in classes-format, see man page) to get a
>>         word-level POS-based model.   Similar with lemma-based LMs
>>         (the lemmas are effectively word classes).
>>
>>
>>     which is the instruction to do this?
>
>     1. You create the class-to-word mapping file (in the format
>     described here
>     <http://www.speech.sri.com/projects/srilm/manpages/classes-format.5.html>)
>     to reflect either your POS-to-word or lemma-to-word mapping.
>     2. Process the training data to replace the words with POS or
>     lemmas, as appropriate.
>     3. Train the ngram portion of the LM by running ngram-count on the
>     training data represented as a sequence of POS tags / lemmas (from
>     step 2).
>
>
>
>>         2. Then interpolate the models using
>>
>>             ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 ....
>>         -lambda ... -mix-lambda2 ... -classes CLASSES
>>
>>         where CLASSES is a classes-format(5) file defining the union
>>         of all the word classes used in the various component models.
>>
>>
>>     to find the lambdas can I use the compute-best-mix, can't I?
>     Exactly.
>
>     Andreas
>
>
>
>
> -- 
> Marta Ruiz Costa-juss?
> martaruizcostajussa at gmail.com <mailto:martaruizcostajussa at gmail.com>
> http://gps-tsc.upc.es/veu/personal/mruiz/mruiz.php3

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130116/6a58b6fb/attachment.html>

From m.shoaie at gmail.com  Thu Jan 17 04:43:39 2013
From: m.shoaie at gmail.com (Maryam Shoaie)
Date: Thu, 17 Jan 2013 16:13:39 +0330
Subject: [SRILM User List] SRILM in windows using Visual C++
Message-ID: <CAFxjT9hA9x0m3+agKO-Ro6yc8E=LNGzRTB4BmM9NP=ha=hSWHQ@mail.gmail.com>

Hi,
I downloaded vc2005 and opened the solution srilm.sln in Visual C++ 2005
express edition. before opening, a window showed this message:
"Solution folders are not supported in this version of visual studio.
Solution folder 'libs' will displayed as unavailable.
Solution folder 'programs' will displayed as unavailable.
some of the properties associated with the solution could not be read."
and when I built it, it does not success. and shows some errors, such as
"no such file or directory" for all .cc files.
On the other hand, when I want to open any of the .cc files, it does not
open and shows this message : "cannot open file"
please help me.
what should I do?
Best Regards,
M. Shoaie
Jan 17
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130117/7e7be662/attachment.html>

From stolcke at icsi.berkeley.edu  Thu Jan 17 09:06:29 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Thu, 17 Jan 2013 09:06:29 -0800
Subject: [SRILM User List] SRILM in windows using Visual C++
In-Reply-To: <CAFxjT9hA9x0m3+agKO-Ro6yc8E=LNGzRTB4BmM9NP=ha=hSWHQ@mail.gmail.com>
References: <CAFxjT9hA9x0m3+agKO-Ro6yc8E=LNGzRTB4BmM9NP=ha=hSWHQ@mail.gmail.com>
Message-ID: <50F82F95.5010906@icsi.berkeley.edu>

On 1/17/2013 4:43 AM, Maryam Shoaie wrote:
> Hi,
> I downloaded vc2005 and opened the solution srilm.sln in Visual C++ 
> 2005 express edition. before opening, a window showed this message:
VS 2010 Express Edition is more recent.  I suggest you try that first.
Andreas


From martaruizcostajussa at gmail.com  Fri Jan 18 05:21:36 2013
From: martaruizcostajussa at gmail.com (Marta Ruiz)
Date: Fri, 18 Jan 2013 21:21:36 +0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <50F76E6C.3020501@icsi.berkeley.edu>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
	<CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
	<50ED1D8F.7000209@icsi.berkeley.edu>
	<CABEBqH+-y+3CO43oqtqGUFzdbUvB1BtOqLFg1aW964--4JCK=w@mail.gmail.com>
	<50F76E6C.3020501@icsi.berkeley.edu>
Message-ID: <CABEBqHKzxteASo4yGXHiYe8M-dRkHiZ47XdV7_Hs0DP2xrXbEw@mail.gmail.com>

The process is killed anyway... Are there any alternatives?

best regards,
Marta

On Thu, Jan 17, 2013 at 11:22 AM, Andreas Stolcke <stolcke at icsi.berkeley.edu
> wrote:

>  On 1/16/2013 6:00 PM, Marta Ruiz wrote:
>
> Hi Andreas,
>
> regarding this issue, I got the error
>
>  class definition has too many fields
>
> That means you must have a very long line in your class definitions file.
> You should have one class membership definition per line.
> If a class has many members you write one per line, for example
>
> NN    cat
> NN    dog
> NN    ball
>
> etc.
>
> Andreas
>
>
> in fact, I wanted to expand a language model of PoS tags into words...
> actually, each PoS has many words related...
>
>
> best regards,
> Marta
>
> On Wed, Jan 9, 2013 at 3:34 PM, Andreas Stolcke <stolcke at icsi.berkeley.edu
> > wrote:
>
>>  On 1/8/2013 6:07 PM, Marta Ruiz wrote:
>>
>> Thanks Andreas, two more questions
>>
>>>
>>> 1. Create a word-based version of each model.  For example, you can
>>> construct a POS-based LM and combine it with a class membership mapping (in
>>> classes-format, see man page) to get a word-level POS-based model.
>>> Similar with lemma-based LMs (the lemmas are effectively word classes).
>>>
>>>
>> which is the instruction to do this?
>>
>>
>>  1. You create the class-to-word mapping file (in the format described
>> here<http://www.speech.sri.com/projects/srilm/manpages/classes-format.5.html>)
>> to reflect either your POS-to-word or lemma-to-word mapping.
>> 2. Process the training data to replace the words with POS or lemmas, as
>> appropriate.
>> 3. Train the ngram portion of the LM by running ngram-count on the
>> training data represented as a sequence of POS tags / lemmas (from step 2).
>>
>>
>>
>>
>>
>>> 2. Then interpolate the models using
>>>
>>>     ngram -bayes 0 -lm LM1 -mix-lm LM2 -mix-lm2 LM3 .... -lambda ...
>>> -mix-lambda2 ... -classes CLASSES
>>>
>>> where CLASSES is a classes-format(5) file defining the union of all the
>>> word classes used in the various component models.
>>>
>>>
>> to find the lambdas can I use the compute-best-mix, can't I?
>>
>>  Exactly.
>>
>> Andreas
>>
>>
>
>
> --
> Marta Ruiz Costa-juss?
> martaruizcostajussa at gmail.com
> http://gps-tsc.upc.es/veu/personal/mruiz/mruiz.php3
>
>
>


-- 
Marta Ruiz Costa-juss?
martaruizcostajussa at gmail.com
http://gps-tsc.upc.es/veu/personal/mruiz/mruiz.php3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130118/98b31ffe/attachment.html>

From stolcke at icsi.berkeley.edu  Fri Jan 18 07:30:57 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Fri, 18 Jan 2013 07:30:57 -0800
Subject: [SRILM User List] linear interpolation of different vocabulary
 language models
In-Reply-To: <CABEBqHKzxteASo4yGXHiYe8M-dRkHiZ47XdV7_Hs0DP2xrXbEw@mail.gmail.com>
References: <CABEBqHKQEABKRX8jWO2BHFHMvWaUQepNKJx=heKA0JZFc=hU9g@mail.gmail.com>
	<50EC7E7D.6050703@icsi.berkeley.edu>
	<CABEBqHJ798PkMfSe_DJ9YLFASabk1S8Wk65nJHZgdtfoJ8tSpQ@mail.gmail.com>
	<50ED1D8F.7000209@icsi.berkeley.edu>
	<CABEBqH+-y+3CO43oqtqGUFzdbUvB1BtOqLFg1aW964--4JCK=w@mail.gmail.com>
	<50F76E6C.3020501@icsi.berkeley.edu>
	<CABEBqHKzxteASo4yGXHiYe8M-dRkHiZ47XdV7_Hs0DP2xrXbEw@mail.gmail.com>
Message-ID: <50F96AB1.4030405@icsi.berkeley.edu>

On 1/18/2013 5:21 AM, Marta Ruiz wrote:
> The process is killed anyway... Are there any alternatives?
That means it's probably working as intended and you're running out of 
memory.
See http://www.speech.sri.com/pipermail/srilm-user/2010q1/000843.html .

Andreas


From chenmengdx at gmail.com  Thu Jan 24 02:24:10 2013
From: chenmengdx at gmail.com (Meng Chen)
Date: Thu, 24 Jan 2013 18:24:10 +0800
Subject: [SRILM User List] How to do 1-best viterbi decoding from lattice?
Message-ID: <CA+bc0mrpqATrpPQXsHKWnd_eK+BOg32CggB7z+tkOptd-VBtvA@mail.gmail.com>

Hi, I want to check the correctness of lattices in HTK format (SLF). So I
need to do the 1-best viterbi decoding from the lattices without loading
new LM. Then compare the hypothesis with standard decoder's 1-pass viterbi
decoding result. My command is as follows:

lattice-tool -debug 2 -in-lattice example.slf -read-htk -old-decoding
-veterbi-decode

However, I found that the hypothesis above is different from the standard
decoder(eg. HTK). Is the command above right?

Thanks!

Meng CHEN
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130124/5de79ef1/attachment.html>

From supersolberg at gmail.com  Thu Jan 24 07:34:02 2013
From: supersolberg at gmail.com (=?ISO-8859-1?Q?Lars_J=F8rgen_Solberg?=)
Date: Thu, 24 Jan 2013 16:34:02 +0100
Subject: [SRILM User List] Did the network protocol change in srilm 1.7?
Message-ID: <5101546A.5060806@gmail.com>

Hi

I have a project where I use the server functionality in srilm, which 
worked fine until i upgraded to version 1.7.

What I do can be outlined as (code snippets are in python):

1 launch a server
$ ngram -server-port 5000 ...

2 connect to the server
conn = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
conn.connect(("localhost", 5000))

3 send a bunch of ngrams separated by newlines
conn.sendall('A A A\nB B B\nC C C\n')

4 recieve the probabilites
reply = conn.recv(4096)

This works fine in version 1.6, but in 1.7 the reply only contains one 
probability.

I guess that one way I can get around this is by only sending one ngram 
at the time, but I would rather not do this as I get increased 
performance when sending and receiving them in chunks.

Is this change in behavior intentional? If so, is there a way of sending 
several ngrams at once to a server?

Regards
Lars J?rgen Solberg

-- 
The moth of wrath goads the rat on! The rat goes berserk!


From stolcke at icsi.berkeley.edu  Thu Jan 24 07:46:40 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Thu, 24 Jan 2013 07:46:40 -0800
Subject: [SRILM User List] How to do 1-best viterbi decoding from
	lattice?
In-Reply-To: <CA+bc0mrpqATrpPQXsHKWnd_eK+BOg32CggB7z+tkOptd-VBtvA@mail.gmail.com>
References: <CA+bc0mrpqATrpPQXsHKWnd_eK+BOg32CggB7z+tkOptd-VBtvA@mail.gmail.com>
Message-ID: <51015760.1040402@icsi.berkeley.edu>

On 1/24/2013 2:24 AM, Meng Chen wrote:
> Hi, I want to check the correctness of lattices in HTK format (SLF). 
> So I need to do the 1-best viterbi decoding from the lattices without 
> loading new LM. Then compare the hypothesis with standard decoder's 
> 1-pass viterbi decoding result. My command is as follows:
>
> lattice-tool -debug 2 -in-lattice example.slf -read-htk -old-decoding 
> -veterbi-decode
>
> However, I found that the hypothesis above is different from the 
> standard decoder(eg. HTK). Is the command above right?
Yes, this is the right command.

You should get the same result with or without -old-decoding.

Consider that the scores (AM, LM, word insertion penalty) might be 
weighted differently in your decoder.

Andreas


From stolcke at icsi.berkeley.edu  Fri Jan 25 00:21:58 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Fri, 25 Jan 2013 00:21:58 -0800
Subject: [SRILM User List] Did the network protocol change in srilm 1.7?
In-Reply-To: <5101546A.5060806@gmail.com>
References: <5101546A.5060806@gmail.com>
Message-ID: <510240A6.1010500@icsi.berkeley.edu>

On 1/24/2013 7:34 AM, Lars J?rgen Solberg wrote:
> Hi
>
> I have a project where I use the server functionality in srilm, which 
> worked fine until i upgraded to version 1.7.
>
> What I do can be outlined as (code snippets are in python):
>
> 1 launch a server
> $ ngram -server-port 5000 ...
>
> 2 connect to the server
> conn = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> conn.connect(("localhost", 5000))
>
> 3 send a bunch of ngrams separated by newlines
> conn.sendall('A A A\nB B B\nC C C\n')
>
> 4 recieve the probabilites
> reply = conn.recv(4096)
>
> This works fine in version 1.6, but in 1.7 the reply only contains one 
> probability.

The very first code implementing the LM server used stdio library calls 
to receive/send data over the network socket.
This happened to allow the behavior you want, but was never intended (or 
documented).  It is also not portable (e.g., to Windows sockets).
When the code was rewritten (before 1.7, actually) to use recv/send 
system calls this no longer worked.

However, I see why you would want to send batches of requests, so this 
seems like a useful thing to support.
I modified the code to do this, and the patch is attached.  Please let 
me know how it works for you.

Andreas

-------------- next part --------------
*** lm/src/LM.cc	4 Dec 2012 20:58:45 -0000	1.92
--- lm/src/LM.cc	25 Jan 2013 08:07:58 -0000	1.93
***************
*** 14,21 ****
--- 14,23 ----
  #include <math.h>
  #include <ctype.h>
  #include <assert.h>
+ #include <string>
  #include "TLSWrapper.h"
  #include "tserror.h"
+ #include "MStringTokUtil.h"
  
  #if !defined(_MSC_VER) && !defined(WIN32)
  #include <unistd.h>
***************
*** 1148,1165 ****
  		exit(-1);
  	    }
  
! 	    char line[REMOTELM_MAXREQUESTLEN + 1];
  	    int msgLen;
  	    unsigned protocolVersion = 1;
  
! 	    while ((msgLen = recv(client, line, sizeof(line)-1, 0)) != SOCKET_ERROR) {
  		if (msgLen == 0) break;
  
! 		line[msgLen] = '\0';
  
  		if (debug(DEBUG_PRINT_WORD_PROBS)) {
  		    dout() << "client " << clientPort << "@" << clientName 
! 			   << ": " << line;
  		}
  
  		VocabString words[maxWordsPerLine + 2];
--- 1150,1178 ----
  		exit(-1);
  	    }
  
! 	    char msgBuffer[REMOTELM_MAXREQUESTLEN + 1];
  	    int msgLen;
  	    unsigned protocolVersion = 1;
  
! 	    while ((msgLen = recv(client, msgBuffer, sizeof(msgBuffer)-1, 0)) != SOCKET_ERROR) {
  		if (msgLen == 0) break;
  
! 		msgBuffer[msgLen] = '\0';
! 		string response = "";
  
+ 		char *strtok_ptr = NULL;
+ 		char *line;
+ 		
+ 		/*
+ 		 * Break message into commands, one per line
+ 		 */
+ 		for (line = MStringTokUtil::strtok_r(msgBuffer, "\n", &strtok_ptr);
+ 		     line != 0;
+ 		     line = MStringTokUtil::strtok_r(0, "\n", &strtok_ptr))
+ 		{
  		    if (debug(DEBUG_PRINT_WORD_PROBS)) {
  			dout() << "client " << clientPort << "@" << clientName 
! 			       << ": " << line << endl;
  		    }
  
  		    VocabString words[maxWordsPerLine + 2];
***************
*** 1252,1266 ****
  			sprintf(outbuf, "%s command unknown\n", REMOTELM_ERROR);
  		    }
  
! 		    if (send(client, outbuf, strlen(outbuf), 0) == SOCKET_ERROR) {
  			cerr << "client " << clientPort << "@" << clientName
  			     << ": send: " << SOCKET_ERROR_STRING << endl;
  			exit(-1);
  		    }
  
  		    if (debug(DEBUG_PRINT_WORD_PROBS)) {
! 			dout() << outbuf;
! 		    }
  		}
  	    }
  
--- 1265,1286 ----
  			    sprintf(outbuf, "%s command unknown\n", REMOTELM_ERROR);
  			}
  
! 			/*
! 			 * Concatenate responses for all commands in the message
! 			 */
! 			response += outbuf;
! 		    }
! 
! 		}
! 
! 		if (send(client, response.c_str(), response.length(), 0) == SOCKET_ERROR) {
  		    cerr << "client " << clientPort << "@" << clientName
  			 << ": send: " << SOCKET_ERROR_STRING << endl;
  		    exit(-1);
  		}
  
  		if (debug(DEBUG_PRINT_WORD_PROBS)) {
! 		    dout() << response;
  		}
  	    }
  

From supersolberg at gmail.com  Fri Jan 25 01:33:52 2013
From: supersolberg at gmail.com (=?windows-1252?Q?Lars_J=F8rgen_Solberg?=)
Date: Fri, 25 Jan 2013 10:33:52 +0100
Subject: [SRILM User List] Did the network protocol change in srilm 1.7?
In-Reply-To: <510240A6.1010500@icsi.berkeley.edu>
References: <5101546A.5060806@gmail.com> <510240A6.1010500@icsi.berkeley.edu>
Message-ID: <51025180.50207@gmail.com>

It works like a charm. Thanks!

-ljs

On 25. jan. 2013 09:21, Andreas Stolcke wrote:
> On 1/24/2013 7:34 AM, Lars J?rgen Solberg wrote:
>> Hi
>>
>> I have a project where I use the server functionality in srilm, which 
>> worked fine until i upgraded to version 1.7.
>>
>> What I do can be outlined as (code snippets are in python):
>>
>> 1 launch a server
>> $ ngram -server-port 5000 ...
>>
>> 2 connect to the server
>> conn = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>> conn.connect(("localhost", 5000))
>>
>> 3 send a bunch of ngrams separated by newlines
>> conn.sendall('A A A\nB B B\nC C C\n')
>>
>> 4 recieve the probabilites
>> reply = conn.recv(4096)
>>
>> This works fine in version 1.6, but in 1.7 the reply only contains 
>> one probability.
>
> The very first code implementing the LM server used stdio library 
> calls to receive/send data over the network socket.
> This happened to allow the behavior you want, but was never intended 
> (or documented).  It is also not portable (e.g., to Windows sockets).
> When the code was rewritten (before 1.7, actually) to use recv/send 
> system calls this no longer worked.
>
> However, I see why you would want to send batches of requests, so this 
> seems like a useful thing to support.
> I modified the code to do this, and the patch is attached.  Please let 
> me know how it works for you.
>
> Andreas
>


-- 
The moth of wrath goads the rat on! The rat goes berserk!


From stolcke at icsi.berkeley.edu  Fri Jan 25 08:17:52 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Fri, 25 Jan 2013 08:17:52 -0800
Subject: [SRILM User List] Did the network protocol change in srilm 1.7?
In-Reply-To: <51025180.50207@gmail.com>
References: <5101546A.5060806@gmail.com> <510240A6.1010500@icsi.berkeley.edu>
	<51025180.50207@gmail.com>
Message-ID: <5102B030.2000501@icsi.berkeley.edu>

On 1/25/2013 1:33 AM, Lars J?rgen Solberg wrote:
> It works like a charm. Thanks!
Glad to hear it.  Please be aware that there is a limit to the string 
length of the message the server will process.
This is the constant REMOTELM_MAXREQUESTLEN in lm/src/RemoteLM.h 
(currently 5000).

Andreas

>
> -ljs
>
> On 25. jan. 2013 09:21, Andreas Stolcke wrote:
>> On 1/24/2013 7:34 AM, Lars J?rgen Solberg wrote:
>>> Hi
>>>
>>> I have a project where I use the server functionality in srilm, 
>>> which worked fine until i upgraded to version 1.7.
>>>
>>> What I do can be outlined as (code snippets are in python):
>>>
>>> 1 launch a server
>>> $ ngram -server-port 5000 ...
>>>
>>> 2 connect to the server
>>> conn = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>>> conn.connect(("localhost", 5000))
>>>
>>> 3 send a bunch of ngrams separated by newlines
>>> conn.sendall('A A A\nB B B\nC C C\n')
>>>
>>> 4 recieve the probabilites
>>> reply = conn.recv(4096)
>>>
>>> This works fine in version 1.6, but in 1.7 the reply only contains 
>>> one probability.
>>
>> The very first code implementing the LM server used stdio library 
>> calls to receive/send data over the network socket.
>> This happened to allow the behavior you want, but was never intended 
>> (or documented).  It is also not portable (e.g., to Windows sockets).
>> When the code was rewritten (before 1.7, actually) to use recv/send 
>> system calls this no longer worked.
>>
>> However, I see why you would want to send batches of requests, so 
>> this seems like a useful thing to support.
>> I modified the code to do this, and the patch is attached. Please let 
>> me know how it works for you.
>>
>> Andreas
>>
>
>


From martaruizcostajussa at gmail.com  Fri Jan 25 23:30:21 2013
From: martaruizcostajussa at gmail.com (Marta Ruiz)
Date: Sat, 26 Jan 2013 15:30:21 +0800
Subject: [SRILM User List] Preliminary CFP: ACL 2013 Second Workshop on
 Hybrid Approaches to Translation (HyTra)
Message-ID: <CABEBqHL98YtxfjUU-B-xiGLqk9bFGq_mYh2-rOupKsAb9jD66Q@mail.gmail.com>

Apologies for multiple postings

Please distribute to colleagues


 =========================================================================


 *Second Workshop on Hybrid Approaches to Translation (HyTra)*


 Co-located with *ACL 2013 <http://acl2013.org/site/>*

Sofia, Bulgaria

August 8, 2013


 Deadline for paper submissions: April 30, 2013


 *http://hytra.barcelonamedia.org/hytra**2013*<http://hytra.barcelonamedia.org/hytra2013>
* *


 =========================================================================




 The Second Workshop on Hybrid Approaches to Translation (HyTra) intends to
further progress on the findings from the first edition which was held
(together with the ESIRMT workshop) as a joint 2-day EACL 2012 workshop.


 The aim of the HyTra workshop is to bring together and share ideas among
researchers developing statistical, example-based, or rule-based
translation systems and who enhance MT systems with elements from the other
approaches. Hereby one relevant focus will be on effectively combining
linguistic and data driven approaches (rule-based and statistical MT).
Another focus will be on hybridization in the context of human translation.


 We solicit contributions including but not limited to the following topics:



   -

   ways and techniques of hybridization
   -

   architectures for the rapid development of hybrid MT systems
   -

   applications of hybrid systems
   -

   hybrid systems dealing with under-resourced languages
   -

   hybrid systems dealing with morphologically rich languages
   -

   using linguistic information (morphology, syntax, semantics) to enhance
   statistical MT (e.g. with hierarchical or factored models)
   -

   using contextual information to enhance statistical MT
   -

   bootstrapping rule-based systems from corpora
   -

   hybrid methods in spoken language translation
   -

   extraction of dictionaries from parallel and comparable corpora
   -

   induction of morphological, grammatical, and translation rules from
   corpora
   -

   machine learning techniques for hybrid MT
   -

   describing structural mappings between languages (e.g. tree-structures
   using synchronous/transduction grammars)
   -

   heuristics for limiting the search space in hybrid MT
   -

   alternative methods for the fair evaluation of the output of different
   types of MT systems (e.g. relying on linguistic criteria)
   -

   system combination approaches such as multi-engine MT (parallel) or
   automatic post-editing (sequential)
   -

   open source tools and free language resources for hybrid MT


 Contributions are to be organized into two tracks, corresponding to short
or long papers. Short paper submission must describe original and
unpublished work without exceeding five pages of content plus one extra
page for references. Characteristics of short papers include: a small,
focused contribution; work in progress; a negative result; an opinion
piece; an interesting application nugget. Long paper submissions must
describe substantial, original, completed and unpublished work without
exceeding eight pages of content plus two extra pages for references.


 *Submission Instructions*


 Authors are invited to submit papers on original and previously
unpublished work. Formatting should be according to ACL 2013 specifications
using LaTeX or MS-Word style files, see *http://acl2013.org/site/call.html*.
Reviewing of papers will be double-blind, so the submissions should not
reveal the authors? identity.


 Submission is electronic in PDF format using the START submission system
at (URL to be announced in subsequent versions of this call)


 Double submission policy: Parallel submission to other meetings or
publications are possible but must be immediately notified to the workshop
contact person (see below).


 Authors of accepted papers will be invited to present their research at
the workshop. The papers will be published in the electronic workshop
proceedings. These will be part of the ACL proceedings, published by ACL.
They will also be made available online via the ACL Anthology.



 *Important Dates*


 April 30, 2013: Paper submissions due

May 24, 2013: Notification of acceptance

Jun 7, 2013: Camera ready papers due

August 8 or 9, 2013: Workshop in Sofia




 *Organizers*


 Marta R. Costa-juss? (Institute for Infocomm Research).

Reinhard Rapp (Universities of Aix-Marseille and Mainz),

Patrik Lambert (Barcelona Media Innovation Center),

Rafael E. Banchs (Institute for Infocomm Research),

Bogdan Babych (University of Leeds)


 *Contact person: *Marta R. Costa-juss? (martaruizcostajussa at gmail.com)


 *Invited Speaker*


 TBA


 *Programme Committee *


 Alexey Baytin, Yandex, Moscow, Russia

N?ria Bel, Universitat Pompeu Fabra, Barcelona, Spain

Pierrette Bouillon, ISSCO/TIM/ETI, University of Geneva, Switzerland

Michael Carl, Copenhagen Business School, Denmark

Marine Carpuat, National Research Council, Canada

Josep Maria Crego, Systran, Paris, France

Oliver Culo, University of Mainz, Germany

Kurt Eberle, Lingenio GmbH, Heidelberg, Germany

Andreas Eisele, DGT (European Commission), Luxembourg

Marcello Federico, Fondazione Bruno Kessler, Trento, Italy

Christian Federmann, Language Technology Lab, DFKI, Saarbr?cken, Germany

Alexander Fraser, University of Stuttgart, Germany

Jos? A. R. Fonollosa, Universitat Polit?cnica de Catalunya, Barcelona, Spain

Tony Hartley, Toyohashi University of Technology, Japan, and University of
Leeds, UK

Maxim Khalilov, TAUS, Amsterdam, The Netherlands

Philipp Koehn, University of Edimburgh, UK

Kevin Knight, University of Southern Carlifornia, US

Udo Kruschwitz, University of Essex, UK

Yanjun Ma, Baidu Inc., Beijing, China

Jos? B. Mari?o, Universitat Polit?cnica de Catalunya, Barcelona, Spain

Maite Melero, Barcelona Media Innovation Center, Barcelona, Spain

Haizhou Li, Institute for Infocomm Research, Singapore

Chris Quirk, Microsoft, USA

Paul Schmidt, Institute for Applied Information Science, Saarbr?cken,
Germany

Anders Sogaard, University of Copenhagen, Denmark

Uta Seewald-Heeg, Anhalt University of Applied Sciences, K?then, Germany

Nasredine Semmar, CEA LIST, Fontenay-aux-Roses, France

Wade Shen, Massachusetts Institute of Technology, Cambridge, USA

Serge Sharoff, University of Leeds, UK

George Tambouratzis, Institute for Language and Speech Processing, Athens,
Greece.

J?rg Tiedemann, University of Uppsala, Sweden

Dekai Wu, The Hong Kong University of Science and Technology, Hong Kong,
China
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130126/7786e129/attachment.html>

From stolcke at icsi.berkeley.edu  Sat Jan 26 09:30:09 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Sat, 26 Jan 2013 09:30:09 -0800
Subject: [SRILM User List] Preliminary CFP: ACL 2013 Second Workshop on
 Hybrid Approaches to Translation (HyTra)
In-Reply-To: <CABEBqHL98YtxfjUU-B-xiGLqk9bFGq_mYh2-rOupKsAb9jD66Q@mail.gmail.com>
References: <CABEBqHL98YtxfjUU-B-xiGLqk9bFGq_mYh2-rOupKsAb9jD66Q@mail.gmail.com>
Message-ID: <510412A1.8060907@icsi.berkeley.edu>

On 1/25/2013 11:30 PM, Marta Ruiz wrote:
>
> Apologies for multiple postings
>
> Please distribute to colleagues
>

Please don't use the srilm-user list for mailings not directly related 
to SRILM.

Thanks

Andreas

>
> =========================================================================
>
>
> *Second Workshop on Hybrid Approaches to Translation (HyTra)*
>
>
> Co-located with _ACL 2013 <http://acl2013.org/site/>_
>
> Sofia, Bulgaria
>
> August 8, 2013
>
>
> Deadline for paper submissions: April 30, 2013
>
>
> _http://hytra.barcelonamedia.org/hytra__2013_ 
> <http://hytra.barcelonamedia.org/hytra2013>__
>
>
> =========================================================================
>
>
>
>
> The Second Workshop on Hybrid Approaches to Translation (HyTra) 
> intends to further progress on the findings from the first edition 
> which was held (together with the ESIRMT workshop) as a joint 2-day 
> EACL 2012 workshop.
>
>
> The aim of the HyTra workshop is to bring together and share ideas 
> among researchers developing statistical, example-based, or rule-based 
> translation systems and who enhance MT systems with elements from the 
> other approaches. Hereby one relevant focus will be on effectively 
> combining linguistic and data driven approaches (rule-based and 
> statistical MT). Another focus will be on hybridization in the context 
> of human translation.
>
>
> We solicit contributions including but not limited to the following 
> topics:
>
>
>  *
>
>     ways and techniques of hybridization
>
>  *
>
>     architectures for the rapid development of hybrid MT systems
>
>  *
>
>     applications of hybrid systems
>
>  *
>
>     hybrid systems dealing with under-resourced languages
>
>  *
>
>     hybrid systems dealing with morphologically rich languages
>
>  *
>
>     using linguistic information (morphology, syntax, semantics) to
>     enhance statistical MT (e.g. with hierarchical or factored models)
>
>  *
>
>     using contextual information to enhance statistical MT
>
>  *
>
>     bootstrapping rule-based systems from corpora
>
>  *
>
>     hybrid methods in spoken language translation
>
>  *
>
>     extraction of dictionaries from parallel and comparable corpora
>
>  *
>
>     induction of morphological, grammatical, and translation rules
>     from corpora
>
>  *
>
>     machine learning techniques for hybrid MT
>
>  *
>
>     describing structural mappings between languages (e.g.
>     tree-structures using synchronous/transduction grammars)
>
>  *
>
>     heuristics for limiting the search space in hybrid MT
>
>  *
>
>     alternative methods for the fair evaluation of the output of
>     different types of MT systems (e.g. relying on linguistic criteria)
>
>  *
>
>     system combination approaches such as multi-engine MT (parallel)
>     or automatic post-editing (sequential)
>
>  *
>
>     open source tools and free language resources for hybrid MT
>
>
> Contributions are to be organized into two tracks, corresponding to 
> short or long papers. Short paper submission must describe original 
> and unpublished work without exceeding five pages of content plus one 
> extra page for references. Characteristics of short papers include: a 
> small, focused contribution; work in progress; a negative result; an 
> opinion piece; an interesting application nugget. Long paper 
> submissions must describe substantial, original, completed and 
> unpublished work without exceeding eight pages of content plus two 
> extra pages for references.
>
>
> *Submission Instructions*
>
>
> Authors are invited to submit papers on original and previously 
> unpublished work. Formatting should be according to ACL 2013 
> specifications using LaTeX or MS-Word style files, see 
> _http://acl2013.org/site/call.html_. Reviewing of papers will be 
> double-blind, so the submissions should not reveal the authors' identity.
>
>
> Submission is electronic in PDF format using the START submission 
> system at (URL to be announced in subsequent versions of this call)
>
>
> Double submission policy: Parallel submission to other meetings or 
> publications are possible but must be immediately notified to the 
> workshop contact person (see below).
>
>
> Authors of accepted papers will be invited to present their research 
> at the workshop. The papers will be published in the electronic 
> workshop proceedings. These will be part of the ACL proceedings, 
> published by ACL. They will also be made available online via the ACL 
> Anthology.
>
>
>
> *Important Dates*
>
>
> April 30, 2013: Paper submissions due
>
> May 24, 2013: Notification of acceptance
>
> Jun 7, 2013: Camera ready papers due
>
> August 8 or 9, 2013: Workshop in Sofia
>
>
>
>
> *Organizers*
>
>
> Marta R. Costa-juss? (Institute for Infocomm Research).
>
> Reinhard Rapp (Universities of Aix-Marseille and Mainz),
>
> Patrik Lambert (Barcelona Media Innovation Center),
>
> Rafael E. Banchs (Institute for Infocomm Research),
>
> Bogdan Babych (University of Leeds)
>
>
> *Contact person: *Marta R. Costa-juss? (martaruizcostajussa at gmail.com 
> <mailto:martaruizcostajussa at gmail.com>)
>
>
> *Invited Speaker*
>
>
> TBA
>
>
> *Programme Committee *
>
>
> Alexey Baytin, Yandex, Moscow, Russia
>
> N?ria Bel, Universitat Pompeu Fabra, Barcelona, Spain
>
> Pierrette Bouillon, ISSCO/TIM/ETI, University of Geneva, Switzerland
>
> Michael Carl, Copenhagen Business School, Denmark
>
> Marine Carpuat, National Research Council, Canada
>
> Josep Maria Crego, Systran, Paris, France
>
> Oliver Culo, University of Mainz, Germany
>
> Kurt Eberle, Lingenio GmbH, Heidelberg, Germany
>
> Andreas Eisele, DGT (European Commission), Luxembourg
>
> Marcello Federico, Fondazione Bruno Kessler, Trento, Italy
>
> Christian Federmann, Language Technology Lab, DFKI, Saarbr?cken, Germany
>
> Alexander Fraser, University of Stuttgart, Germany
>
> Jos? A. R. Fonollosa, Universitat Polit?cnica de Catalunya, Barcelona, 
> Spain
>
> Tony Hartley, Toyohashi University of Technology, Japan, and 
> University of Leeds, UK
>
> Maxim Khalilov, TAUS, Amsterdam, The Netherlands
>
> Philipp Koehn, University of Edimburgh, UK
>
> Kevin Knight, University of Southern Carlifornia, US
>
> Udo Kruschwitz, University of Essex, UK
>
> Yanjun Ma, Baidu Inc., Beijing, China
>
> Jos? B. Mari?o, Universitat Polit?cnica de Catalunya, Barcelona, Spain
>
> Maite Melero, Barcelona Media Innovation Center, Barcelona, Spain
>
> Haizhou Li, Institute for Infocomm Research, Singapore
>
> Chris Quirk, Microsoft, USA
>
> Paul Schmidt, Institute for Applied Information Science, Saarbr?cken, 
> Germany
>
> Anders Sogaard, University of Copenhagen, Denmark
>
> Uta Seewald-Heeg, Anhalt University of Applied Sciences, K?then, Germany
>
> Nasredine Semmar, CEA LIST, Fontenay-aux-Roses, France
>
> Wade Shen, Massachusetts Institute of Technology, Cambridge, USA
>
> Serge Sharoff, University of Leeds, UK
>
> George Tambouratzis, Institute for Language and Speech Processing, 
> Athens, Greece.
>
> J?rg Tiedemann, University of Uppsala, Sweden
>
> Dekai Wu, The Hong Kong University of Science and Technology, Hong 
> Kong, China
>
>
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130126/2242c5a5/attachment.html>

From stolcke at icsi.berkeley.edu  Mon Jan 28 13:47:34 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Mon, 28 Jan 2013 13:47:34 -0800
Subject: [SRILM User List] Question on unsmoothed estimates
In-Reply-To: <CAOUwyXtDpn5rELowMWM2qucW0kWntB3NcoXz8cj4VKLo2ptGeA@mail.gmail.com>
References: <CAOUwyXtDpn5rELowMWM2qucW0kWntB3NcoXz8cj4VKLo2ptGeA@mail.gmail.com>
Message-ID: <5106F1F6.6060102@icsi.berkeley.edu>

On 1/28/2013 1:38 PM, Avneesh Saluja wrote:
> Hello Andreas,
>
> I hope you're doing well.  I have a quick question on SRILM and its 
> ability to compute completely unsmoothed probability estimates.  Of 
> course, I can use the counts output of ngram-count and then compute 
> probabilities from there, but since ngram-count already does this, I 
> thought I should use that facility, but I'm not able to get it to do 
> what I want.
>
> Here's an example, with a small LM consisting of a training corpus of 
> only 12,534 words (using "wc" on the file).  There are 1872 unigrams 
> (as per the LM output).  The exact command I used to generate my LM is:
>
>  ~/tools/srilm/bin/i686-m64/ngram-count -order 3 -text 
> ../data/lm_training/small/train.txt -cdiscount 0 -lm unsmoothed-lm
>
> First, I see that the word "accident" occurs 5 times in my corpus. 
>  Therefore, one would expect the unigram probability to be 
> log10(5/12534) = -3.40.  However, the result in SRILM is -3.45, 
> indicating some sort of smoothing going on.
The end-of-sentence tokens also count as events in the model.  So your 
denominator is larger than you assume, hence the lower probability estimate.

If you use ngram-count -debug 4 you will see exactly what quantities go 
into the estimation of each ngram probability.

>
> Furthermore, when looking at higher order n-grams, I see that there 
> are only 2 trigrams where the first two words are "hilton hotel" --> 
> "hilton hotel ?" and "hilton hotel ,", the count of the former is 2 
> and the count of the latter is 1.  However, in the resulting 
> unsmoothed n-gram, I only see the former entry "hilton hotel ?", and 
> it has the right log probability (-0.176 --> 10^(-0.176) = 0.67), but 
> I can't find the entry "hilton hotel ,", which should have a log 
> probability of log10(1/3) = -0.477.  However, for another instance, 
> say the bigrams w_1, w_2 where w_1 = "twelve", I get the correct 
> probability estimates for the bigrams.
By default trigrams (and 4grams, etc. ) that occur only once are omitted 
from the LM.  Use -gt3min 1 to change that.

Andreas


From stolcke at icsi.berkeley.edu  Tue Feb 12 12:05:48 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Tue, 12 Feb 2013 12:05:48 -0800
Subject: [SRILM User List] Fw: SRI help please
In-Reply-To: <1360689684.66935.YahooMailNeo@web28805.mail.ir2.yahoo.com>
References: <1360688851.35618.YahooMailNeo@web28804.mail.ir2.yahoo.com>
	<1360689684.66935.YahooMailNeo@web28805.mail.ir2.yahoo.com>
Message-ID: <511AA09C.8090007@icsi.berkeley.edu>

Noor,

please submit your questions to srilm-user at speech.sri.com in the future 
(after joining the list).

On 2/12/2013 9:21 AM, Noor Noor wrote:
>
> Dear Sir,
>
> I want to interpolate LM, I followed this steps:
>
> 1. ngram -debug 2 -ppl text1.txt > L1.txt
> ngram -debug 2 -ppl text2.txt > L2.txt
>
> 2. ngram-class  -text text1.txt -class-counts 3 -classes cls1 
> -numclasses 2
>   ngram-class  -text text2.txt -class-counts 3 -classes cls2 
> -numclasses 2
> to get allcls
>
> 3. How I can use compute-best-mix to combine L1.txt L2.txt and how get 
> the result
>  Can write to me the right command to my input to get LP output ?

You run
         compute-best-mix L1.txt L2.txt
and use the first number after "best lambda" in the output as the 
-lambda value for ngram.

>
> 4. how I can combine cls1 and cls2?
You need to rename the class labels in the cls1 and cls2 files to not 
clash.  It might be easier (and actually work better) to infer classes 
from the union of text1.txt and text2.txt, and then use those classes in 
both component LMs.

Also, you don't mention the step of training the class-based LMs once 
you have the class definitions.  You need to use something like

             replace-words-with-classes classes=allcls text1.txt | \
             ngram-count -text - -lm LM1 ....

to build your class-based LMs.

> 5. Is the following command the right one to interpolate LM?
> /ngram -bayes 0 -lm LM1 -mix-lm LM2 -lambda //LP////-mix-lambda2 ... 
> -classes allcls/
> /
> /
> /what //-mix-lambda2 correct value for my input?/

With only two models to interpolate, the command would be

     ngram -bayes 0 -lm LM1 -mix-lm LM2 -lambda LP  -classes allcls

The second lambda value is inferred from LP since they have to add up to 1.

Andreas

> /
> /
> /
> /
> /please kindly can help me in this matter?/
> /
> /
> /Best Regards,/
> /
> /
> /Noor/
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130212/2658c60d/attachment.html>

From is13-noreply at inria.fr  Thu Feb 21 13:52:33 2013
From: is13-noreply at inria.fr (Interspeech 2013 - Third announcement)
Date: Thu, 21 Feb 2013 22:52:33 +0100
Subject: [SRILM User List] Interspeech 2013, Lyon (France),
 25-29 August - Third call for papers and participation
Message-ID: <51269721.3040603@inria.fr>

INTERSPEECH 2013 Conference
with special focus on "Speech in Life Sciences and Human Societies"
25-29 August 2013, Lyon (France)
Third announcement

--- Flash news ---

**Submission Deadline: March 18, 2013**
Paper submission system is now open
Instruction for authors : 
http://dev.europa-organisation.com/2013/interspeech/content/paper-submission-procedure 



--- Welcome message ---

Dear colleagues,

In a few months, the 2013 edition of the Interspeech Conference will 
start. We look forward to welcoming you in Lyon, France.

In the meantime, a new version of the conference website has been 
launched to provide you with extended information to prepare your 
participation to the event next August: http://www.interspeech2013.org

As of today, prospective authors are formally invited to prepare their 
paper submissions.

The Interspeech conferences reflect the dynamism of our scientific 
community, which has steadily strengthened over the years.

We are looking forward to receiving, from all areas of the world, an 
abundant flow of high-quality contributions on the wide range of topics 
covered by the conference.

Fr?d?ric Bimbot (chair), C?cile Fougeron (co-chair), Fran?ois Pellegrino 
(co-chair)


--- Important Dates and Links ---

Calls for Satellite Workshops, Tutorials and Special Sessions are now 
closed

Call for Papers

Submission Deadline: March 18, 2013
Notification of acceptance : May 22, 2013
Camera-ready paper : May 29, 2013

List of conference areas : 
http://dev.europa-organisation.com/2013/interspeech/content/conference-areas
Instruction for authors : 
http://dev.europa-organisation.com/2013/interspeech/content/paper-submission-procedure 

Download the Author's kit: 
http://dev.europa-organisation.com/2013/interspeech/sites/default/files/IS2013_paperkit.zip

Call for Show & Tell and Other Special Events
Submission Deadline: April 19, 2013

Conference Dates: 25-29 August 2013


--- General Information ---

Paper format

 From this year on, a new paper format will be used: 4 pages of text 
plus one page (maximum) for references only. This will facilitate the 
sometimes difficult arbitration between work description and citation 
coverage.

Paper submission procedure

Authors will be asked to determine to which topic(s) their contributions 
relates (see list of areas and topics on the website) and/or if they 
wish to contribute to a special session (see list below). Under the 
coordination of the Technical Programme Chairs (Lori Lamel and Pascal 
Perrier) and with the supervision of a pool of 28 area chairs, each 
paper proposal will be reviewed by at least three specialists of the 
area to which it belongs. Papers must be submitted via the on-line paper 
submission system, accessible through the conference website. The 
deadline for submitting a paper is March 18th, 2013.

Registration fees

The definitive fees have not been fixed yet, but our target is to be 
within the range of previous Interspeech editions (that is 520 ? in 
Florence, 470 ? in Portland, for ISCA member, full-delegate early 
registration). The fees will be announced about a week before the paper 
submission deadline.

Student/retired participants

Students will benefit from reduced rates (-50% or more) and will be 
offered the possibility to opt for a student accommodation package at an 
attractive cost (typically, in the range of 200-250 ? for 5 nights). A 
significant number of grants will be offered, to further reduce the 
financial burden for student (or retired) attendees, when needed. Keep 
posted for more info.


--- Special sessions proposal ---

Special sessions at Interspeech are intended to bring together 
researchers in relevant fields of interest, either on a targeted topic, 
or beyond the traditional scope of the conference.

For Interspeech 2013, eleven proposals of special sessions are currently 
pending.

Prospective authors are offered the option to relate their paper 
proposal to one of these sessions. All papers relating to a special 
session will follow the regular reviewing process, by anonymous and 
independent reviewers.

Only special sessions with a sufficient number of accepted papers will 
ultimately be scheduled in the final
program. All other accepted papers will be presented in an adequate 
regular session.

To consult the list of special session proposals :
http://dev.europa-organisation.com/2013/interspeech/content/special-sessions

The final list of special sessions will be determined and published 
after the review process, on May 22, 2013.


--- Information about satellite workshops ---

We are happy to announce that over a dozen satellite workshops are 
organized before or after the Interspeech conference.

This is an opportunity to participate to specific scientific events 
related to current "hot topics" in speech science.

These workshops take place in locations with easy access to/from Lyon 
(Grenoble, Paris, Aix-en-Provence, Barcelona and more?).

More information on satellite workshops can be found on the Interspeech 
web site :
http://dev.europa-organisation.com/2013/interspeech/content/satellite-workshops


--- Become a sponsor of the conference ---

If you want to become a sponsor or an exhibitor at the conference :
http://dev.europa-organisation.com/2013/interspeech/content/become-partner-interspeech-2013


--- Contacts ---

Conference Chairs: conference-chairs at interspeech2013.org
Technical Programme Chairs: technical-chairs at interspeech2013.org
Organizers: organizers at interspeech2013.org
Webmaster: webmaster at interspeech2013.org


From noor2004noor at ymail.com  Sat Feb 23 20:54:18 2013
From: noor2004noor at ymail.com (Noor Noor)
Date: Sun, 24 Feb 2013 04:54:18 +0000 (GMT)
Subject: [SRILM User List] srilm problem in sphin3
Message-ID: <1361681658.82642.YahooMailNeo@web28805.mail.ir2.yahoo.com>

Dear Sir,

I have problem in sphinx 3 when I create LM using SRILM tools.

I used?

ngram-count -text text1.txt -lm LM


then I try to use?

lm_convert (sphinx3 command) to convert the LM to DMP file.

I get this error:
Fail to read inputfn LM in inputfmt TXT


Also I try to use

lm3g2dmp also I have the same problem

the error is "line 385: No \data\ mark in LM file"


also I try to use?add-dummy-bows?
my question is how I can used arpa file that I create it from SRILM tools in sphinx 3 testing? is that possible or not?
how I can resolve this?problem?

Thanks and I look forward to hearing from you as soon as possible

Regards,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130224/d1a15a2c/attachment.html>

From stolcke at icsi.berkeley.edu  Sat Feb 23 21:53:44 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Sat, 23 Feb 2013 21:53:44 -0800
Subject: [SRILM User List] srilm problem in sphin3
In-Reply-To: <1361681658.82642.YahooMailNeo@web28805.mail.ir2.yahoo.com>
References: <1361681658.82642.YahooMailNeo@web28805.mail.ir2.yahoo.com>
Message-ID: <5129AAE8.6060601@icsi.berkeley.edu>

I don't know much about Sphinx, but this old post might have helpful 
information:

http://www.speech.sri.com/pipermail/srilm-user/2008q2/000645.html

Andreas

On 2/23/2013 8:54 PM, Noor Noor wrote:
> Dear Sir,
>
> I have problem in sphinx 3 when I create LM using SRILM tools.
>
> I used
>
> ngram-count -text text1.txt -lm LM
>
> then I try to use
>
> lm_convert (sphinx3 command) to convert the LM to DMP file.
> I get this error:
> Fail to read inputfn LM in inputfmt TXT
>
> Also I try to use
>
> lm3g2dmp also I have the same problem
> the error is "line 385: No \data\ mark in LM file"
>
> also I try to use add-dummy-bows
> my question is how I can used arpa file that I create it from SRILM 
> tools in sphinx 3 testing? is that possible or not?
> how I can resolve this problem?
>
> Thanks and I look forward to hearing from you as soon as possible
>
> Regards,
>
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130223/eb3d0bf4/attachment.html>

From noor2004noor at ymail.com  Sat Feb 23 23:45:09 2013
From: noor2004noor at ymail.com (Noor Noor)
Date: Sun, 24 Feb 2013 07:45:09 +0000 (GMT)
Subject: [SRILM User List] Fw:  srilm problem in sphin3
In-Reply-To: <1361689040.41929.YahooMailNeo@web28806.mail.ir2.yahoo.com>
References: <1361681658.82642.YahooMailNeo@web28805.mail.ir2.yahoo.com>
	<51299FEF.6070706@cs.cmu.edu>
	<1361689040.41929.YahooMailNeo@web28806.mail.ir2.yahoo.com>
Message-ID: <1361691909.10425.YahooMailNeo@web28805.mail.ir2.yahoo.com>




Dear Mr. Long,

Thanks a lot.

About your email:

Do you mean use sphinx 3 like this command using LM.arpa?

decode -cepdir mfc -ctl file.ctl -hyp out.trn -dict file.dict
-fdict file.filler -lm LM.arpa -hmm HMMpath - test.log

can do that or need another parameter to support this command for arpa file?

what I did:

1. create LM using srilm order 3 gram2.?add-dummy-bows? ?3.?sort-lm?
4. testing LM in sphinx 3 as above.

but still I have the following problem:

LM is not a DMP file .....?lm_read_advance(LM.arpa, 9.500000e+00, 7.000000e-01, 7.000000e-01 74961 [Arbitrary Fmt], Weighted Apply) failed


Please tell me what is the problem here?

Thanks and best regards,

Noor


________________________________
 From: Long Qin <lqin+ at cs.cmu.edu>
To: Noor Noor <noor2004noor at ymail.com> 
Cc: "srilm-user at speech.sri.com" <srilm-user at speech.sri.com> 
Sent: Saturday, 23 February 2013, 21:06
Subject: Re: [SRILM User List] srilm problem in sphin3
 
Hi Noor,

You don't have to convert the LM to DMP format. Sphinx3 can directly 
load ARPA LM in TXT format. But you will still need to run 
add-dummy-bows to add 0 to ngrams whose back-off weight is zero.

Sometimes, the bigrams or trigrams are not sorted in the same order as 
unigram, which may cause an error in Sphinx3. If that happens, you will 
need to re-sort ngrams in your LM.

Thanks,
Long

On 2/23/13 11:54 PM, Noor Noor wrote:
> Dear Sir,
>
> I have problem in sphinx 3 when I create LM using SRILM tools.
>
> I used
>
> ngram-count -text text1.txt -lm LM
>
> then I try to use
>
> lm_convert (sphinx3 command) to convert the LM to DMP file.
> I get this
 error:
> Fail to read inputfn LM in inputfmt TXT
>
> Also I try to use
>
> lm3g2dmp also I have the same problem
> the error is "line 385: No \data\ mark in LM file"
>
> also I try to use add-dummy-bows
> my question is how I can used arpa file that I create it from SRILM
> tools in sphinx 3 testing? is that possible or not?
> how I can resolve this problem?
>
> Thanks and I look forward to hearing from you as soon as possible
>
> Regards,
>
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130224/7ffbaf43/attachment.html>

From yangyiycc at gmail.com  Wed Feb 27 07:42:41 2013
From: yangyiycc at gmail.com (Yi Yang)
Date: Wed, 27 Feb 2013 10:42:41 -0500
Subject: [SRILM User List] Srilm C++ library compile problem
Message-ID: <CAKNTL4ipP4VSA=RAcv_M+mEcgLLBKsOjY3T6niPi-LrHGs9RRg@mail.gmail.com>

Hi,

I have a problem when I try to compile my C++ problem with srilm library.

My commend is

g++ srilmTest.cc -I/nethome/yyang319/software/srilm/include/
-L/nethome/yyang319/software/srilm/lib/i686-m64/ -ldstruct -lflm -lmisc
-loolm -ldl -lpthread -o srilmTest


And I got a lot of this kind of errors:

/nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
undefined reference to `srilm_tls_get_key()'
/nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o):/nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
more undefined references to `srilm_tls_get_key()' follow


/nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): In
function `LHash<unsigned int, NBestWordInfo>::alloc(unsigned int)':
/nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:140:
undefined reference to `BM_malloc'
/nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): In
function `LHash<unsigned int, NBestWordInfo>::insert(unsigned int, bool&)':
/nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:407:
undefined reference to `BM_free'


I tried to address it for a long time, but failed.

Does anybody have solutions for this problem?

Thanks a lot!

-- 
Sincerely,
Yi Yang
http://www.cc.gatech.edu/~yyang319/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130227/fed81dde/attachment.html>

From stolcke at icsi.berkeley.edu  Wed Feb 27 14:19:32 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Wed, 27 Feb 2013 14:19:32 -0800
Subject: [SRILM User List] Srilm C++ library compile problem
In-Reply-To: <CAKNTL4ipP4VSA=RAcv_M+mEcgLLBKsOjY3T6niPi-LrHGs9RRg@mail.gmail.com>
References: <CAKNTL4ipP4VSA=RAcv_M+mEcgLLBKsOjY3T6niPi-LrHGs9RRg@mail.gmail.com>
Message-ID: <512E8674.80409@icsi.berkeley.edu>

On 2/27/2013 7:42 AM, Yi Yang wrote:
> Hi,
>
> I have a problem when I try to compile my C++ problem with srilm library.
>
> My commend is
>
> g++ srilmTest.cc -I/nethome/yyang319/software/srilm/include/ 
> -L/nethome/yyang319/software/srilm/lib/i686-m64/ -ldstruct -lflm 
> -lmisc -loolm -ldl -lpthread -o srilmTest

Did SRIM itself build correctly?  Have you run '"make test" and did it work?
>
>
> And I got a lot of this kind of errors:
>
> /nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50: 
> undefined reference to `srilm_tls_get_key()'
> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o):/nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50: 
> more undefined references to `srilm_tls_get_key()' follow

If you are having trouble with TLS you should try compiling with -DNO_TLS .

>
>
> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): 
> In function `LHash<unsigned int, NBestWordInfo>::alloc(unsigned int)':
> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:140: 
> undefined reference to `BM_malloc'
> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): 
> In function `LHash<unsigned int, NBestWordInfo>::insert(unsigned int, 
> bool&)':
> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:407: 
> undefined reference to `BM_free'
This strongly indicates that your libraries didn't all build properly 
(hence my question above).
BM_malloc, BM_free are defined in -ldstruct .

Andreas

>
>
> I tried to address it for a long time, but failed.
>
> Does anybody have solutions for this problem?
>
> Thanks a lot!
>
> -- 
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/~yyang319/ 
> <http://www.cc.gatech.edu/%7Eyyang319/>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130227/edfa7f70/attachment.html>

From yangyiycc at gmail.com  Wed Feb 27 17:03:07 2013
From: yangyiycc at gmail.com (Yi Yang)
Date: Wed, 27 Feb 2013 20:03:07 -0500
Subject: [SRILM User List] Srilm C++ library compile problem
In-Reply-To: <512E8674.80409@icsi.berkeley.edu>
References: <CAKNTL4ipP4VSA=RAcv_M+mEcgLLBKsOjY3T6niPi-LrHGs9RRg@mail.gmail.com>
	<512E8674.80409@icsi.berkeley.edu>
Message-ID: <CAKNTL4jPjUGeZh4N5g4rnVBwCPyTe7DraVWHsp29GoWOcCFJKg@mail.gmail.com>

Hi Andreas,

Thanks for your reply.

I rebuilt SRILM again. My machine is 64bit Linux. The process is as
following:

1.  set NO_TCL = X in common/Makefile.machine.i686-m64
2. make World
3. make MACHINE_TYPE=i686-m64 World
4. make test
5. make cleanest

And after I run make test, the last lines of the output are:

*** Running test lattice-word-posteriors ***

real 0m0.033s
user 0m0.006s
sys 0m0.023s
lattice-word-posteriors: stdout output IDENTICAL.
lattice-word-posteriors: stderr output IDENTICAL.
make[1]: Leaving directory `/nethome/yyang319/software/srilm/lattice/test'


I did not meet any errors during the building process. However, I still got
the same errors when I try to compile my program with srilm library.

Did I make some mistakes while building SRILM?

Thank you so much.
Yi




On Wed, Feb 27, 2013 at 5:19 PM, Andreas Stolcke
<stolcke at icsi.berkeley.edu>wrote:

>  On 2/27/2013 7:42 AM, Yi Yang wrote:
>
> Hi,
>
>  I have a problem when I try to compile my C++ problem with srilm library.
>
>  My commend is
>
>  g++ srilmTest.cc -I/nethome/yyang319/software/srilm/include/
> -L/nethome/yyang319/software/srilm/lib/i686-m64/ -ldstruct -lflm -lmisc
> -loolm -ldl -lpthread -o srilmTest
>
>
> Did SRIM itself build correctly?  Have you run '"make test" and did it
> work?
>
>
>
>  And I got a lot of this kind of errors:
>
>  /nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
> undefined reference to `srilm_tls_get_key()'
> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o):/nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
> more undefined references to `srilm_tls_get_key()' follow
>
>
> If you are having trouble with TLS you should try compiling with -DNO_TLS .
>
>
>
>
>  /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): In
> function `LHash<unsigned int, NBestWordInfo>::alloc(unsigned int)':
> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:140:
> undefined reference to `BM_malloc'
> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): In
> function `LHash<unsigned int, NBestWordInfo>::insert(unsigned int, bool&)':
> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:407:
> undefined reference to `BM_free'
>
> This strongly indicates that your libraries didn't all build properly
> (hence my question above).
> BM_malloc, BM_free are defined in -ldstruct .
>
> Andreas
>
>
>
>  I tried to address it for a long time, but failed.
>
>  Does anybody have solutions for this problem?
>
>  Thanks a lot!
>
>  --
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/~yyang319/
>
>
> _______________________________________________
> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>


-- 
Sincerely,
Yi Yang
http://www.cc.gatech.edu/~yyang319/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130227/dfba448f/attachment.html>

From yangyiycc at gmail.com  Wed Feb 27 17:11:20 2013
From: yangyiycc at gmail.com (Yi Yang)
Date: Wed, 27 Feb 2013 20:11:20 -0500
Subject: [SRILM User List] Srilm C++ library compile problem
In-Reply-To: <CAKNTL4jPjUGeZh4N5g4rnVBwCPyTe7DraVWHsp29GoWOcCFJKg@mail.gmail.com>
References: <CAKNTL4ipP4VSA=RAcv_M+mEcgLLBKsOjY3T6niPi-LrHGs9RRg@mail.gmail.com>
	<512E8674.80409@icsi.berkeley.edu>
	<CAKNTL4jPjUGeZh4N5g4rnVBwCPyTe7DraVWHsp29GoWOcCFJKg@mail.gmail.com>
Message-ID: <CAKNTL4jwLQ6Q_=bX_O8GtBq_+5DbxbGUsfh+REt=gvAWFYHBFw@mail.gmail.com>

Hi Andreas,

I just addressed this problem. What I need to do is just changing the order
of "-ldstruct -lflm -lmisc -loolm -ldl -lpthread" ...

I change it to  "-loolm -ldstruct -lflm -lmisc -ldl -lpthread" and it works
...


Thank you very much for your help.

Best,
Yi


On Wed, Feb 27, 2013 at 8:03 PM, Yi Yang <yangyiycc at gmail.com> wrote:

> Hi Andreas,
>
> Thanks for your reply.
>
> I rebuilt SRILM again. My machine is 64bit Linux. The process is as
> following:
>
> 1.  set NO_TCL = X in common/Makefile.machine.i686-m64
> 2. make World
> 3. make MACHINE_TYPE=i686-m64 World
> 4. make test
> 5. make cleanest
>
> And after I run make test, the last lines of the output are:
>
> *** Running test lattice-word-posteriors ***
>
> real 0m0.033s
> user 0m0.006s
> sys 0m0.023s
> lattice-word-posteriors: stdout output IDENTICAL.
> lattice-word-posteriors: stderr output IDENTICAL.
> make[1]: Leaving directory `/nethome/yyang319/software/srilm/lattice/test'
>
>
> I did not meet any errors during the building process. However, I still
> got the same errors when I try to compile my program with srilm library.
>
> Did I make some mistakes while building SRILM?
>
> Thank you so much.
> Yi
>
>
>
>
> On Wed, Feb 27, 2013 at 5:19 PM, Andreas Stolcke <
> stolcke at icsi.berkeley.edu> wrote:
>
>>  On 2/27/2013 7:42 AM, Yi Yang wrote:
>>
>> Hi,
>>
>>  I have a problem when I try to compile my C++ problem with srilm
>> library.
>>
>>  My commend is
>>
>>  g++ srilmTest.cc -I/nethome/yyang319/software/srilm/include/
>> -L/nethome/yyang319/software/srilm/lib/i686-m64/ -ldstruct -lflm -lmisc
>> -loolm -ldl -lpthread -o srilmTest
>>
>>
>> Did SRIM itself build correctly?  Have you run '"make test" and did it
>> work?
>>
>>
>>
>>  And I got a lot of this kind of errors:
>>
>>  /nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
>> undefined reference to `srilm_tls_get_key()'
>> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o):/nethome/yyang319/software/srilm/lm/src/../../include/TLSWrapper.h:50:
>> more undefined references to `srilm_tls_get_key()' follow
>>
>>
>> If you are having trouble with TLS you should try compiling with -DNO_TLS
>> .
>>
>>
>>
>>
>>  /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o):
>> In function `LHash<unsigned int, NBestWordInfo>::alloc(unsigned int)':
>> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:140:
>> undefined reference to `BM_malloc'
>> /nethome/yyang319/software/srilm/lib/i686-m64//liboolm.a(WordMesh.o): In
>> function `LHash<unsigned int, NBestWordInfo>::insert(unsigned int, bool&)':
>> /nethome/yyang319/software/srilm/lm/src/../../include/LHash.cc:407:
>> undefined reference to `BM_free'
>>
>> This strongly indicates that your libraries didn't all build properly
>> (hence my question above).
>> BM_malloc, BM_free are defined in -ldstruct .
>>
>> Andreas
>>
>>
>>
>>  I tried to address it for a long time, but failed.
>>
>>  Does anybody have solutions for this problem?
>>
>>  Thanks a lot!
>>
>>  --
>> Sincerely,
>> Yi Yang
>> http://www.cc.gatech.edu/~yyang319/
>>
>>
>> _______________________________________________
>> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>>
>
>
> --
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/~yyang319/
>



-- 
Sincerely,
Yi Yang
http://www.cc.gatech.edu/~yyang319/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130227/b505db85/attachment.html>

From mohsen.jadidi at gmail.com  Sun Mar  3 13:46:20 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Sun, 3 Mar 2013 22:46:20 +0100
Subject: [SRILM User List] installation on mac lion error on negram class
Message-ID: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>

Hello everybody,

I am trying to install versiobn 1.7 on mac, but it keeps giving me error:
.
.
.
.
 ./fngram-count.cc ./fngram.cc | sed -e "s&^\([^
]\)&../obj/macosx"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >>
Dependencies.macosx
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
/Users/mohsenjadidi/Downloads/srilm/sbin/generate-program-dependencies
../bin/macosx ../obj/macosx "" pngram fngram-count fngram  | sed -e
"s&\.o&.o&g" >> Dependencies.macosx
rm -f Dependencies.macosx
c++ -Wreturn-type -Wimplicit -DINSTANTIATE_TEMPLATES   -I/usr/include -I.
-I../../include -DHAVE_ZOPEN -MM  ./Lattice.cc ./LatticeAlign.cc
./LatticeExpand.cc ./LatticeIndex.cc ./LatticeNBest.cc ./LatticeNgrams.cc
./LatticeReduce.cc ./HTKLattice.cc ./LatticeLM.cc ./LatticeThreads.cc
./LatticeDecode.cc ./testLattice.cc ./lattice-tool.cc | sed -e "s&^\([^
]\)&../obj/macosx"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >>
Dependencies.macosx
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
cc1plus: warning: command line option '-Wimplicit' is valid for C/ObjC but
not for C++ [enabled by default]
/Users/mohsenjadidi/Downloads/srilm/sbin/generate-program-dependencies
../bin/macosx ../obj/macosx "" testLattice  lattice-tool | sed -e
"s&\.o&.o&g" >> Dependencies.macosx
rm -f Dependencies.macosx
/Users/mohsenjadidi/Downloads/srilm/sbin/generate-program-dependencies
../bin/macosx ../obj/macosx ""  | sed -e "s&\.o&.o&g" >> Dependencies.macosx
make release-libraries
for subdir in misc dstruct lm flm lattice utils; do \
(cd $subdir/src; make SRILM=/Users/mohsenjadidi/Downloads/srilm
MACHINE_TYPE=macosx OPTION= MAKE_PIC= release-libraries) || exit 1; \
 done
make[2]: Nothing to be done for `release-libraries'.
make[2]: Nothing to be done for `release-libraries'.
make[2]: Nothing to be done for `release-libraries'.
make[2]: Nothing to be done for `release-libraries'.
make[2]: Nothing to be done for `release-libraries'.
make[2]: Nothing to be done for `release-libraries'.
make release-programs
for subdir in misc dstruct lm flm lattice utils; do \
(cd $subdir/src; make SRILM=/Users/mohsenjadidi/Downloads/srilm
MACHINE_TYPE=macosx OPTION= MAKE_PIC= release-programs) || exit 1; \
 done
make[2]: Nothing to be done for `release-programs'.
make[2]: Nothing to be done for `release-programs'.
c++ -Wreturn-type -Wimplicit -DINSTANTIATE_TEMPLATES   -I/usr/include -I.
-I../../include -DHAVE_ZOPEN   -u _matherr -L../../lib/macosx  -g -O2
-fno-common -o ../bin/macosx/ngram ../obj/macosx/ngram.o
../obj/macosx/liboolm.a  ../../lib/macosx/libflm.a
../../lib/macosx/libdstruct.a ../../lib/macosx/libmisc.a -L/usr/lib -ltcl
-lm  -liconv 2>&1 | c++filt
Undefined symbols for architecture x86_64:
  "std::ctype<char>::_M_widen_init() const", referenced from:
      makeLoglinearMixLM(Array<char const*>, Vocab&, SubVocab*, unsigned
int, LM*, Array<double>) in ngram.o
      LHash<unsigned int, unsigned int*>::dump() const in
liboolm.a(MultiwordVocab.o)
      LHash<char const*, unsigned int*>::dump() const in
liboolm.a(RefList.o)
      ClassNgram::prefixProb(unsigned int, unsigned int const*, float&,
TextStats&) in liboolm.a(ClassNgram.o)
      ClassNgram::expand(unsigned int, unsigned int) in
liboolm.a(ClassNgram.o)
      NgramCounts<float>::readBinaryNode(Trie<unsigned int, float>&,
unsigned int, unsigned int, File&, long long&, bool, Array<unsigned int>&)
in liboolm.a(ClassNgram.o)
      NgramCounts<float>::readBinary(File&, unsigned int, bool) in
liboolm.a(ClassNgram.o)
      ...
  "_libiconv", referenced from:
      File::fgetsUTF8(char*, int) in libmisc.a(File.o)
  "_libiconv_close", referenced from:
      File::~File() in libmisc.a(File.o)
      File::~File() in libmisc.a(File.o)
      File::reopen(char const*, char const*) in libmisc.a(File.o)
      File::reopen(char const*) in libmisc.a(File.o)
      File::reopen(char const*, unsigned long, int) in libmisc.a(File.o)
      File::reopen(std::basic_string<char, std::char_traits<char>,
std::allocator<char> >&, int) in libmisc.a(File.o)
  "_libiconv_open", referenced from:
      File::fgetsUTF8(char*, int) in libmisc.a(File.o)
ld: symbol(s) not found for architecture x86_64
collect2: error: ld returned 1 exit status
test -f ../bin/macosx/ngram
make[2]: *** [../bin/macosx/ngram] Error 1
make[1]: *** [release-programs] Error 1
make: *** [World] Error 2








Why it's happening? I managed to install the  version 1.6 without any
problem but I couln't find file ngram-count?

I ran the make file both with and without specifying the platform and wi
Also I tried changing the gcc flag in common.Make..macosx from c++ to gcc
but didnt work.

Any help is appreciated
-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130303/d5e1f877/attachment.html>

From stolcke at icsi.berkeley.edu  Sun Mar  3 19:31:19 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Sun, 03 Mar 2013 19:31:19 -0800
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
Message-ID: <51341587.6000001@icsi.berkeley.edu>

On 3/3/2013 1:46 PM, mohsen jadidi wrote:
> Hello everybody,
>
> I am trying to install versiobn 1.7 on mac, but it keeps giving me error:
The cc1plus warning should disappear if you change 
common/Makefile.machine.macosx  to have

    GCC_FLAGS = -Wreturn-type
    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES

The link error seems to have something to do with the iconv library not 
being properly installed for 64bit builds.
You might be able to work around this by building with

     make NO_ICONV=1  ...

I hope someone with MacOSX savvy can provide a way to make the linking 
with -liconv actually work.

Andreas



From mohsen.jadidi at gmail.com  Mon Mar  4 10:40:49 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Mon, 4 Mar 2013 19:40:49 +0100
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <51341587.6000001@icsi.berkeley.edu>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
Message-ID: <CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>

Thanks but it didnt fixed the problem! Just removed the warnings


On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke
<stolcke at icsi.berkeley.edu>wrote:

> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>
>> Hello everybody,
>>
>> I am trying to install versiobn 1.7 on mac, but it keeps giving me error:
>>
> The cc1plus warning should disappear if you change
> common/Makefile.machine.macosx  to have
>
>    GCC_FLAGS = -Wreturn-type
>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>
> The link error seems to have something to do with the iconv library not
> being properly installed for 64bit builds.
> You might be able to work around this by building with
>
>     make NO_ICONV=1  ...
>
> I hope someone with MacOSX savvy can provide a way to make the linking
> with -liconv actually work.
>
> Andreas
>
>
>


-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130304/9eb65708/attachment.html>

From kutlak.roman at gmail.com  Mon Mar  4 11:20:17 2013
From: kutlak.roman at gmail.com (Roman Kutlak)
Date: Mon, 4 Mar 2013 19:20:17 +0000
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
Message-ID: <7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>

You should check that you have libiconv in /usr/lib and that you have the 64bit version using the following command:

file /usr/lib/libiconv.dylib 

You should see something like the following:
/usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
/usr/lib/libiconv.dylib (for architecture i386):	Mach-O dynamically linked shared library i386
/usr/lib/libiconv.dylib (for architecture x86_64):	Mach-O 64-bit dynamically linked shared library x86_64

Roman

On 4 Mar 2013, at 18:40, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:

> Thanks but it didnt fixed the problem! Just removed the warnings 
> 
> 
> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <stolcke at icsi.berkeley.edu> wrote:
> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
> Hello everybody,
> 
> I am trying to install versiobn 1.7 on mac, but it keeps giving me error:
> The cc1plus warning should disappear if you change common/Makefile.machine.macosx  to have
> 
>    GCC_FLAGS = -Wreturn-type
>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
> 
> The link error seems to have something to do with the iconv library not being properly installed for 64bit builds.
> You might be able to work around this by building with
> 
>     make NO_ICONV=1  ...
> 
> I hope someone with MacOSX savvy can provide a way to make the linking with -liconv actually work.
> 
> Andreas
> 
> 
> 
> 
> 
> -- 
> Mohsen Jadidi
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130304/38692708/attachment.html>

From mohsen.jadidi at gmail.com  Mon Mar  4 12:35:30 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Mon, 4 Mar 2013 21:35:30 +0100
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
	<7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
Message-ID: <CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>

this is mine:

/usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
/usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
dynamically linked shared library x86_64
/usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically linked
shared library i386


On Mon, Mar 4, 2013 at 8:20 PM, Roman Kutlak <kutlak.roman at gmail.com> wrote:

> You should check that you have libiconv in /usr/lib and that you have the
> 64bit version using the following command:
>
> file /usr/lib/libiconv.dylib
>
> You should see something like the following:
> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
> linked shared library i386
> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
> dynamically linked shared library x86_64
>
> Roman
>
> On 4 Mar 2013, at 18:40, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>
> Thanks but it didnt fixed the problem! Just removed the warnings
>
>
> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <stolcke at icsi.berkeley.edu
> > wrote:
>
>> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>>
>>> Hello everybody,
>>>
>>> I am trying to install versiobn 1.7 on mac, but it keeps giving me error:
>>>
>> The cc1plus warning should disappear if you change
>> common/Makefile.machine.macosx  to have
>>
>>    GCC_FLAGS = -Wreturn-type
>>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>>
>> The link error seems to have something to do with the iconv library not
>> being properly installed for 64bit builds.
>> You might be able to work around this by building with
>>
>>     make NO_ICONV=1  ...
>>
>> I hope someone with MacOSX savvy can provide a way to make the linking
>> with -liconv actually work.
>>
>> Andreas
>>
>>
>>
>
>
> --
> Mohsen Jadidi
>  _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>



-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130304/5a94fff0/attachment.html>

From wooters at icsi.berkeley.edu  Mon Mar  4 15:26:47 2013
From: wooters at icsi.berkeley.edu (Charles Wooters)
Date: Mon, 4 Mar 2013 15:26:47 -0800
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
	<7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
	<CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>
Message-ID: <CAAbXKqcx4vqYR7NQVxEELOUYb9JrSQHsZ5VjUFvjiz_QSZP_cg@mail.gmail.com>

I was able to compile v1.7 on my macbook pro (2.6 GHz Intel Core i7)
running OSX 10.8.2 following these instructions:

http://www1.icsi.berkeley.edu/~wooters/SRILM/index.html

-Chuck


---
Chuck Wooters
International Computer Science Institute
http://www.icsi.berkeley.edu


On Mon, Mar 4, 2013 at 12:35 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:

> this is mine:
>
> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
> dynamically linked shared library x86_64
> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
> linked shared library i386
>
>
> On Mon, Mar 4, 2013 at 8:20 PM, Roman Kutlak <kutlak.roman at gmail.com>wrote:
>
>> You should check that you have libiconv in /usr/lib and that you have the
>> 64bit version using the following command:
>>
>> file /usr/lib/libiconv.dylib
>>
>>  You should see something like the following:
>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
>> linked shared library i386
>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
>> dynamically linked shared library x86_64
>>
>> Roman
>>
>> On 4 Mar 2013, at 18:40, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>>
>> Thanks but it didnt fixed the problem! Just removed the warnings
>>
>>
>> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <
>> stolcke at icsi.berkeley.edu> wrote:
>>
>>> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>>>
>>>> Hello everybody,
>>>>
>>>> I am trying to install versiobn 1.7 on mac, but it keeps giving me
>>>> error:
>>>>
>>> The cc1plus warning should disappear if you change
>>> common/Makefile.machine.macosx  to have
>>>
>>>    GCC_FLAGS = -Wreturn-type
>>>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>>>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>>>
>>> The link error seems to have something to do with the iconv library not
>>> being properly installed for 64bit builds.
>>> You might be able to work around this by building with
>>>
>>>     make NO_ICONV=1  ...
>>>
>>> I hope someone with MacOSX savvy can provide a way to make the linking
>>> with -liconv actually work.
>>>
>>> Andreas
>>>
>>>
>>>
>>
>>
>> --
>> Mohsen Jadidi
>>  _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>
>
>
> --
> Mohsen Jadidi
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130304/bd30260e/attachment.html>

From mohsen.jadidi at gmail.com  Mon Mar  4 22:46:54 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Tue, 5 Mar 2013 07:46:54 +0100
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAAbXKqcx4vqYR7NQVxEELOUYb9JrSQHsZ5VjUFvjiz_QSZP_cg@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
	<7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
	<CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>
	<CAAbXKqcx4vqYR7NQVxEELOUYb9JrSQHsZ5VjUFvjiz_QSZP_cg@mail.gmail.com>
Message-ID: <CAOfo52AQCkuvV313j7WJdv3w0UvqM+yESQH-FqMfV3r6M+N2AQ@mail.gmail.com>

did you install any of prerequisites libraries?


On Tue, Mar 5, 2013 at 12:26 AM, Charles Wooters
<wooters at icsi.berkeley.edu>wrote:

> I was able to compile v1.7 on my macbook pro (2.6 GHz Intel Core i7)
> running OSX 10.8.2 following these instructions:
>
> http://www1.icsi.berkeley.edu/~wooters/SRILM/index.html
>
> -Chuck
>
>
> ---
> Chuck Wooters
> International Computer Science Institute
> http://www.icsi.berkeley.edu
>
>
> On Mon, Mar 4, 2013 at 12:35 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>
>> this is mine:
>>
>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
>> dynamically linked shared library x86_64
>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
>> linked shared library i386
>>
>>
>> On Mon, Mar 4, 2013 at 8:20 PM, Roman Kutlak <kutlak.roman at gmail.com>wrote:
>>
>>> You should check that you have libiconv in /usr/lib and that you have
>>> the 64bit version using the following command:
>>>
>>> file /usr/lib/libiconv.dylib
>>>
>>>  You should see something like the following:
>>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
>>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
>>> linked shared library i386
>>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
>>> dynamically linked shared library x86_64
>>>
>>> Roman
>>>
>>> On 4 Mar 2013, at 18:40, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>>>
>>> Thanks but it didnt fixed the problem! Just removed the warnings
>>>
>>>
>>> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <
>>> stolcke at icsi.berkeley.edu> wrote:
>>>
>>>> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>>>>
>>>>> Hello everybody,
>>>>>
>>>>> I am trying to install versiobn 1.7 on mac, but it keeps giving me
>>>>> error:
>>>>>
>>>> The cc1plus warning should disappear if you change
>>>> common/Makefile.machine.macosx  to have
>>>>
>>>>    GCC_FLAGS = -Wreturn-type
>>>>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>>>>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>>>>
>>>> The link error seems to have something to do with the iconv library not
>>>> being properly installed for 64bit builds.
>>>> You might be able to work around this by building with
>>>>
>>>>     make NO_ICONV=1  ...
>>>>
>>>> I hope someone with MacOSX savvy can provide a way to make the linking
>>>> with -liconv actually work.
>>>>
>>>> Andreas
>>>>
>>>>
>>>>
>>>
>>>
>>> --
>>> Mohsen Jadidi
>>>  _______________________________________________
>>> SRILM-User site list
>>> SRILM-User at speech.sri.com
>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>
>>>
>>>
>>> _______________________________________________
>>> SRILM-User site list
>>> SRILM-User at speech.sri.com
>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>
>>
>>
>>
>> --
>> Mohsen Jadidi
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>
>


-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130305/7eefc5d5/attachment.html>

From wooters at icsi.berkeley.edu  Tue Mar  5 08:52:25 2013
From: wooters at icsi.berkeley.edu (Charles Wooters)
Date: Tue, 5 Mar 2013 08:52:25 -0800
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAOfo52AQCkuvV313j7WJdv3w0UvqM+yESQH-FqMfV3r6M+N2AQ@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
	<7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
	<CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>
	<CAAbXKqcx4vqYR7NQVxEELOUYb9JrSQHsZ5VjUFvjiz_QSZP_cg@mail.gmail.com>
	<CAOfo52AQCkuvV313j7WJdv3w0UvqM+yESQH-FqMfV3r6M+N2AQ@mail.gmail.com>
Message-ID: <CAAbXKqfnw6=bx8oKiuOri4UCaqk7DrrDtm38Qb_CE=xxnNN0MA@mail.gmail.com>

As for prerequisites, I installed these two using the homebrew package
manager (http://mxcl.github.com/homebrew/):

- gawk (this also installed xz)
- gzip

These were already installed on my machine (perhaps by Xcode or Xcode
command-line utilities):

 - make
 - tcsh
 - tcl
 - bzip2

This is the only one that I haven't installed (but I will now):

 - p7zip

Hope that helps.

-Chuck


---
Chuck Wooters
International Computer Science Institute
http://www.icsi.berkeley.edu


On Mon, Mar 4, 2013 at 10:46 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:

> did you install any of prerequisites libraries?
>
>
> On Tue, Mar 5, 2013 at 12:26 AM, Charles Wooters <
> wooters at icsi.berkeley.edu> wrote:
>
>> I was able to compile v1.7 on my macbook pro (2.6 GHz Intel Core i7)
>> running OSX 10.8.2 following these instructions:
>>
>> http://www1.icsi.berkeley.edu/~wooters/SRILM/index.html
>>
>> -Chuck
>>
>>
>> ---
>> Chuck Wooters
>> International Computer Science Institute
>> http://www.icsi.berkeley.edu
>>
>>
>> On Mon, Mar 4, 2013 at 12:35 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>>
>>> this is mine:
>>>
>>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
>>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
>>> dynamically linked shared library x86_64
>>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
>>> linked shared library i386
>>>
>>>
>>> On Mon, Mar 4, 2013 at 8:20 PM, Roman Kutlak <kutlak.roman at gmail.com>wrote:
>>>
>>>> You should check that you have libiconv in /usr/lib and that you have
>>>> the 64bit version using the following command:
>>>>
>>>> file /usr/lib/libiconv.dylib
>>>>
>>>>  You should see something like the following:
>>>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2 architectures
>>>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O dynamically
>>>> linked shared library i386
>>>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O 64-bit
>>>> dynamically linked shared library x86_64
>>>>
>>>> Roman
>>>>
>>>> On 4 Mar 2013, at 18:40, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>>>>
>>>> Thanks but it didnt fixed the problem! Just removed the warnings
>>>>
>>>>
>>>> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <
>>>> stolcke at icsi.berkeley.edu> wrote:
>>>>
>>>>> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>>>>>
>>>>>> Hello everybody,
>>>>>>
>>>>>> I am trying to install versiobn 1.7 on mac, but it keeps giving me
>>>>>> error:
>>>>>>
>>>>> The cc1plus warning should disappear if you change
>>>>> common/Makefile.machine.macosx  to have
>>>>>
>>>>>    GCC_FLAGS = -Wreturn-type
>>>>>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>>>>>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>>>>>
>>>>> The link error seems to have something to do with the iconv library
>>>>> not being properly installed for 64bit builds.
>>>>> You might be able to work around this by building with
>>>>>
>>>>>     make NO_ICONV=1  ...
>>>>>
>>>>> I hope someone with MacOSX savvy can provide a way to make the linking
>>>>> with -liconv actually work.
>>>>>
>>>>> Andreas
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Mohsen Jadidi
>>>>  _______________________________________________
>>>> SRILM-User site list
>>>> SRILM-User at speech.sri.com
>>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> SRILM-User site list
>>>> SRILM-User at speech.sri.com
>>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>>
>>>
>>>
>>>
>>> --
>>> Mohsen Jadidi
>>>
>>> _______________________________________________
>>> SRILM-User site list
>>> SRILM-User at speech.sri.com
>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>
>>
>>
>
>
> --
> Mohsen Jadidi
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130305/c84370de/attachment.html>

From wooters at icsi.berkeley.edu  Wed Mar  6 09:45:16 2013
From: wooters at icsi.berkeley.edu (Charles Wooters)
Date: Wed, 6 Mar 2013 09:45:16 -0800
Subject: [SRILM User List] maxent models using libLBFGS on Mac OSX
Message-ID: <CAAbXKqcoHDSsdFtRdg0rKVAoRqKAj-Wq--6dcHsEPw52xEBSnw@mail.gmail.com>

If any Mac users out there are interested in building maximum entropy
models with the SRILM Toolkit (using the libLBFGS optimization library as
described in the 'INSTALL' file), I have written up some instructions about
how to compile libLBFGS on the Mac (including the specifics about how to
add it in to the SRILM Toolkit compile process.)

It is a simple process, but it took a little time to figure out, so I
thought I'd post this in case it might help someone out.

You can find the instructions here:

https://gist.github.com/wooters/5095712

-Chuck

---
Chuck Wooters
International Computer Science Institute
http://www.icsi.berkeley.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130306/c449ddf8/attachment.html>

From stolcke at icsi.berkeley.edu  Wed Mar  6 19:23:42 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Wed, 06 Mar 2013 19:23:42 -0800
Subject: [SRILM User List] maxent models using libLBFGS on Mac OSX
In-Reply-To: <CAAbXKqcoHDSsdFtRdg0rKVAoRqKAj-Wq--6dcHsEPw52xEBSnw@mail.gmail.com>
References: <CAAbXKqcoHDSsdFtRdg0rKVAoRqKAj-Wq--6dcHsEPw52xEBSnw@mail.gmail.com>
Message-ID: <5138083E.3080103@icsi.berkeley.edu>

Chuck,

your information on MacOS building is very helpful!  Thanks.

I just wanted to clarify that the maxent functionality is a contribution 
from Tanel Alumae that was orginally available from his own web site, 
but is now fully integrated into SRILM.  Thanks Tanel!   It is part of 
the current 1.7.1-beta version, and it would be helpful if more people 
could test it on a variety of platforms.

Andreas

On 3/6/2013 9:45 AM, Charles Wooters wrote:
> If any Mac users out there are interested in building maximum entropy 
> models with the SRILM Toolkit (using the libLBFGS optimization library 
> as described in the 'INSTALL' file), I have written up some 
> instructions about how to compile libLBFGS on the Mac (including the 
> specifics about how to add it in to the SRILM Toolkit compile process.)
>
> It is a simple process, but it took a little time to figure out, so I 
> thought I'd post this in case it might help someone out.
>
> You can find the instructions here:
>
> https://gist.github.com/wooters/5095712
>
> -Chuck
>
> ---
> Chuck Wooters
> International Computer Science Institute
> http://www.icsi.berkeley.edu
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130306/06551ab1/attachment.html>

From mohsen.jadidi at gmail.com  Thu Mar  7 12:23:22 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Thu, 7 Mar 2013 21:23:22 +0100
Subject: [SRILM User List] c++ sample for building language model
Message-ID: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>

Hey,

I need to use srilm in my c++ code to build ML. all the examples and slides
in the internet explained it using ngram-count command not code. I know
should use <Ngram.h> <Vocab.h>. can you point me to a starting point?

cheers,


-- 
Mohsen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130307/747fbcda/attachment.html>

From alumae at gmail.com  Fri Mar  8 00:34:49 2013
From: alumae at gmail.com (=?ISO-8859-1?Q?Tanel_Alum=E4e?=)
Date: Fri, 8 Mar 2013 10:34:49 +0200
Subject: [SRILM User List] maxent models using libLBFGS on Mac OSX
In-Reply-To: <5138083E.3080103@icsi.berkeley.edu>
References: <CAAbXKqcoHDSsdFtRdg0rKVAoRqKAj-Wq--6dcHsEPw52xEBSnw@mail.gmail.com>
	<5138083E.3080103@icsi.berkeley.edu>
Message-ID: <CACP=pzKVhdtQcAzwJ7ij8+Y5K1rVXQw=Sz02ut9x0Z2n_Y1woQ@mail.gmail.com>

Hello!

The documentation on how to enable, compile and use the maxent
functionality is given here:
http://www.phon.ioc.ee/~tanela/srilm-me/

Hope this helps,
Tanel


On Thu, Mar 7, 2013 at 5:23 AM, Andreas Stolcke
<stolcke at icsi.berkeley.edu>wrote:

>  Chuck,
>
> your information on MacOS building is very helpful!  Thanks.
>
> I just wanted to clarify that the maxent functionality is a contribution
> from Tanel Alumae that was orginally available from his own web site, but
> is now fully integrated into SRILM.  Thanks Tanel!   It is part of the
> current 1.7.1-beta version, and it would be helpful if more people could
> test it on a variety of platforms.
>
> Andreas
>
>
> On 3/6/2013 9:45 AM, Charles Wooters wrote:
>
>  If any Mac users out there are interested in building maximum entropy
> models with the SRILM Toolkit (using the libLBFGS optimization library as
> described in the 'INSTALL' file), I have written up some instructions about
> how to compile libLBFGS on the Mac (including the specifics about how to
> add it in to the SRILM Toolkit compile process.)
>
>  It is a simple process, but it took a little time to figure out, so I
> thought I'd post this in case it might help someone out.
>
>  You can find the instructions here:
>
>  https://gist.github.com/wooters/5095712
>
>  -Chuck
>
>   ---
> Chuck Wooters
> International Computer Science Institute
>  http://www.icsi.berkeley.edu
>
>
> _______________________________________________
> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130308/a64fc0f4/attachment.html>

From yangyiycc at gmail.com  Fri Mar  8 08:06:57 2013
From: yangyiycc at gmail.com (Yi Yang)
Date: Fri, 8 Mar 2013 11:06:57 -0500
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
Message-ID: <CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>

Hi Mohsen,

Hope the following codes can be helpful:

void SrilmTest::srilm_init(const char* fname, int order) {
  File file(fname, "r", 0);
  assert(file);
  ngram = new Ngram(vocab, order);
  ngram->read(file, false);
  cerr << "Done\n";
}

int SrilmTest::srilm_getvoc(const char* word) {
  return vocab.getIndex((VocabString)word);
}

float SrilmTest::srilm_wordprob(int w, int* context) {
  return (float)ngram->wordProb(w, (VocabIndex*)context);
}


On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:

> Hey,
>
> I need to use srilm in my c++ code to build ML. all the examples and
> slides in the internet explained it using ngram-count command not code. I
> know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>
> cheers,
>
>
> --
> Mohsen
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>



-- 
Sincerely,
Yi Yang
http://www.cc.gatech.edu/~yyang319/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130308/d62eff95/attachment.html>

From stolcke at icsi.berkeley.edu  Fri Mar  8 12:37:35 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Fri, 08 Mar 2013 12:37:35 -0800
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
Message-ID: <513A4C0F.4070104@icsi.berkeley.edu>

On 3/8/2013 8:06 AM, Yi Yang wrote:
> Hi Mohsen,
>
> Hope the following codes can be helpful:
you forgot to create the Vocab object.  In your case you could create it 
globally to your code works otherwise:

Vocab vocab;

Andreas

>
> void SrilmTest::srilm_init(const char* fname, int order) {
>   File file(fname, "r", 0);
>   assert(file);

>   ngram = new Ngram(vocab, order);
>   ngram->read(file, false);
>   cerr << "Done\n";
> }
>
> int SrilmTest::srilm_getvoc(const char* word) {
>   return vocab.getIndex((VocabString)word);
> }
>
> float SrilmTest::srilm_wordprob(int w, int* context) {
>   return (float)ngram->wordProb(w, (VocabIndex*)context);
> }
>
>
> On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com 
> <mailto:mohsen.jadidi at gmail.com>> wrote:
>
>     Hey,
>
>     I need to use srilm in my c++ code to build ML. all the examples
>     and slides in the internet explained it using ngram-count command
>     not code. I know should use <Ngram.h> <Vocab.h>. can you point me
>     to a starting point?
>
>     cheers,
>
>
>     -- 
>     Mohsen
>
>     _______________________________________________
>     SRILM-User site list
>     SRILM-User at speech.sri.com <mailto:SRILM-User at speech.sri.com>
>     http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
>
> -- 
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/~yyang319/ 
> <http://www.cc.gatech.edu/%7Eyyang319/>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130308/2102b7d3/attachment.html>

From yhifny at yahoo.com  Fri Mar  8 12:49:53 2013
From: yhifny at yahoo.com (yasser hifny)
Date: Fri, 8 Mar 2013 12:49:53 -0800 (PST)
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <513A4C0F.4070104@icsi.berkeley.edu>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
Message-ID: <1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>

Hi Mohsen,

code sample ?that I use in my work is

// gloable variables
Vocab*g_srilm_vocab;

Ngram*g_model;

double GetLogWordProb(const std::string& strCurrWord,const vector<std::string>& vstrHistory)
{
floatfResult;

size_t len = 1 + vstrHistory.size()+1;
VocabIndex* WordIDs=new VocabIndex[len];
WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()), g_srilm_vocab->unkIndex());
for(size_t i=0;i<vstrHistory.size();i++)
WordIDs[i+1] = vstrHistory[i]!="<s>"?
g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()), g_srilm_vocab->unkIndex()):Vocab_None;
WordIDs[vstrHistory.size()+1] = Vocab_None;


for(size_t k=0; k<len;k++)
{
DEBUG("k=%d wordindex:%d wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
}
fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
if(fResult==LogP_Zero)fResult=0.0;
DEBUG("prob=%f",fResult);

//g_model->sentenceProb(words, stats);
delete[] WordIDs;

return fResult;

}


//in the ?main function


//--------------------------------
// Load LM
//--------------------------------
g_srilm_vocab= new Vocab;
g_model=new Ngram(*g_srilm_vocab,nOrder);
File file(strLangModelFile.c_str(), "r");
if(!file)
{
ERROR("Could not open file %s",strLangModelFile.c_str());
}
g_model->read(file, 0);
for(size_t i=0;i<nOrder;i++)
TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));


Best regards,
Yasser


________________________________
 From: Andreas Stolcke <stolcke at icsi.berkeley.edu>
To: Yi Yang <yangyiycc at gmail.com> 
Cc: SRILM-User at speech.sri.com 
Sent: Friday, March 8, 2013 10:37 PM
Subject: Re: [SRILM User List] c++ sample for building language model
 

On 3/8/2013 8:06 AM, Yi Yang wrote:

Hi Mohsen, 
>
>
>Hope the following codes can be helpful:
you forgot to create the Vocab object.? In your case you could create it globally to your code works otherwise:

Vocab vocab;

Andreas



>
>void SrilmTest::srilm_init(const char* fname, int order) {
>? File file(fname, "r", 0);
>? assert(file);

? ngram = new Ngram(vocab, order);
>? ngram->read(file, false);
>? cerr << "Done\n";
>}
>
>
>int SrilmTest::srilm_getvoc(const char* word) {
>? return vocab.getIndex((VocabString)word);
>}
>
>
>float SrilmTest::srilm_wordprob(int w, int* context) {
>? return (float)ngram->wordProb(w, (VocabIndex*)context);
>}
>
>
>
>On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>
>Hey,? 
>>
>>
>>I need to use srilm in my c++ code to build ML. all the examples and slides in the internet explained it using ngram-count command not code. I know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>>
>>
>>cheers,
>>
>>
>>
-- 
>>Mohsen
>>
>>_______________________________________________
>>SRILM-User site list
>>SRILM-User at speech.sri.com
>>http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>
>
>
>
-- 
>Sincerely,
>Yi Yang
>http://www.cc.gatech.edu/%7Eyyang319/
>
>
>
>_______________________________________________
SRILM-User site list SRILM-User at speech.sri.com http://www.speech.sri.com/mailman/listinfo/srilm-user

_______________________________________________
SRILM-User site list
SRILM-User at speech.sri.com
http://www.speech.sri.com/mailman/listinfo/srilm-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130308/679cf313/attachment.html>

From mohsen.jadidi at gmail.com  Fri Mar  8 13:10:52 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Fri, 8 Mar 2013 22:10:52 +0100
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
	<1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
Message-ID: <CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>

Thanks for your reply. Very helpful. But don't we need to use ssIndex and
seIndex to build lexicon language model  and then use it for next step?




On Fri, Mar 8, 2013 at 9:49 PM, yasser hifny <yhifny at yahoo.com> wrote:

> Hi Mohsen,
>
> code sample  that I use in my work is
>
> // gloable variables
> Vocab* g_srilm_vocab;
> Ngram* g_model;
>
> double GetLogWordProb(const std::string& strCurrWord,const
> vector<std::string>& vstrHistory)
> {
> float fResult;
>
>  size_t len = 1 + vstrHistory.size()+1;
> VocabIndex* WordIDs=new VocabIndex[len];
>  WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()),
> g_srilm_vocab->unkIndex());
> for(size_t i=0;i<vstrHistory.size();i++)
> WordIDs[i+1] = vstrHistory[i]!="<s>"?
> g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()),
> g_srilm_vocab->unkIndex()):Vocab_None;
>  WordIDs[vstrHistory.size()+1] = Vocab_None;
>
>
>  for(size_t k=0; k<len;k++)
> {
> DEBUG("k=%d wordindex:%d
> wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
> }
> fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
>  if(fResult==LogP_Zero)fResult=0.0;
> DEBUG("prob=%f",fResult);
>
> //g_model->sentenceProb(words, stats);
> delete[] WordIDs;
>
> return fResult;
>
> }
>
>
> //in the  main function
>
>
> //--------------------------------
>  // Load LM
> //--------------------------------
>  g_srilm_vocab= new Vocab;
> g_model=new Ngram(*g_srilm_vocab,nOrder);
>  File file(strLangModelFile.c_str(), "r");
>  if(!file)
> {
>  ERROR("Could not open file %s",strLangModelFile.c_str());
>  }
> g_model->read(file, 0);
>  for(size_t i=0;i<nOrder;i++)
> TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));
>
>
> Best regards,
> Yasser
>
>   ------------------------------
> *From:* Andreas Stolcke <stolcke at icsi.berkeley.edu>
> *To:* Yi Yang <yangyiycc at gmail.com>
> *Cc:* SRILM-User at speech.sri.com
> *Sent:* Friday, March 8, 2013 10:37 PM
> *Subject:* Re: [SRILM User List] c++ sample for building language model
>
>  On 3/8/2013 8:06 AM, Yi Yang wrote:
>
> Hi Mohsen,
>
>  Hope the following codes can be helpful:
>
> you forgot to create the Vocab object.  In your case you could create it
> globally to your code works otherwise:
>
> Vocab vocab;
>
> Andreas
>
>
>  void SrilmTest::srilm_init(const char* fname, int order) {
>   File file(fname, "r", 0);
>   assert(file);
>
>
>     ngram = new Ngram(vocab, order);
>   ngram->read(file, false);
>   cerr << "Done\n";
> }
>
>  int SrilmTest::srilm_getvoc(const char* word) {
>   return vocab.getIndex((VocabString)word);
> }
>
>  float SrilmTest::srilm_wordprob(int w, int* context) {
>   return (float)ngram->wordProb(w, (VocabIndex*)context);
>  }
>
>
> On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>
> Hey,
>
>  I need to use srilm in my c++ code to build ML. all the examples and
> slides in the internet explained it using ngram-count command not code. I
> know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>
>  cheers,
>
>
>  --
> Mohsen
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
>
>  --
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/%7Eyyang319/
>
>
> _______________________________________________
> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>



-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130308/c048457d/attachment.html>

From sgetachew92 at yahoo.com  Sat Mar  9 14:33:23 2013
From: sgetachew92 at yahoo.com (Solomon Getachew)
Date: Sat, 9 Mar 2013 14:33:23 -0800 (PST)
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
	<1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
	<CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
Message-ID: <1362868403.15307.YahooMailNeo@web126204.mail.ne1.yahoo.com>

Dear All?
?I will like?developed Multiple pronunciation Modeling in ASR for Amharic language based on Knowledge Based i need to get sample code is there any one help me?
thanks in advance?


________________________________
 From: mohsen jadidi <mohsen.jadidi at gmail.com>
To: yasser hifny <yhifny at yahoo.com> 
Cc: "SRILM-User at speech.sri.com" <SRILM-User at speech.sri.com> 
Sent: Friday, March 8, 2013 1:10 PM
Subject: Re: [SRILM User List] c++ sample for building language model
 

Thanks for your reply. Very helpful. But don't we need to use ssIndex and seIndex to build lexicon language model ?and then use it for next step??





On Fri, Mar 8, 2013 at 9:49 PM, yasser hifny <yhifny at yahoo.com> wrote:

Hi Mohsen,
>
>
>code sample ?that I use in my work is
>
>
>// gloable variables
>Vocab*g_srilm_vocab;
>
>Ngram*g_model;
>
>
>double GetLogWordProb(const std::string& strCurrWord,const vector<std::string>& vstrHistory)
>{
>floatfResult;
>
>
>size_t len = 1 + vstrHistory.size()+1;
>VocabIndex* WordIDs=new VocabIndex[len];
>WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()), g_srilm_vocab->unkIndex());
>for(size_t i=0;i<vstrHistory.size();i++)
>WordIDs[i+1] = vstrHistory[i]!="<s>"?
>g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()), g_srilm_vocab->unkIndex()):Vocab_None;
>WordIDs[vstrHistory.size()+1] = Vocab_None;
>
>
>
>
>for(size_t k=0; k<len;k++)
>{
>DEBUG("k=%d wordindex:%d wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
>}
>fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
>if(fResult==LogP_Zero)fResult=0.0;
>DEBUG("prob=%f",fResult);
>
>
>//g_model->sentenceProb(words, stats);
>delete[] WordIDs;
>
>
>return fResult;
>
>
>}
>
>
>
>
>//in the ?main function
>
>
>
>
>//--------------------------------
>// Load LM
>//--------------------------------
>g_srilm_vocab= new Vocab;
>g_model=new Ngram(*g_srilm_vocab,nOrder);
>File file(strLangModelFile.c_str(), "r");
>if(!file)
>{
>ERROR("Could not open file %s",strLangModelFile.c_str());
>}
>g_model->read(file, 0);
>for(size_t i=0;i<nOrder;i++)
>TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));
>
>
>
>
>Best regards,
>Yasser
>
>
>
>________________________________
> From: Andreas Stolcke <stolcke at icsi.berkeley.edu>
>To: Yi Yang <yangyiycc at gmail.com> 
>Cc: SRILM-User at speech.sri.com 
>Sent: Friday, March 8, 2013 10:37 PM
>Subject: Re: [SRILM User List] c++ sample for building language model
> 
>
>On 3/8/2013 8:06 AM, Yi Yang wrote:
>
>Hi Mohsen, 
>>
>>
>>Hope the following codes can be helpful:
you forgot to create the Vocab object.? In your case you could create it globally to your code works otherwise:
>
>Vocab vocab;
>
>Andreas
>
>
>
>>
>>void SrilmTest::srilm_init(const char* fname, int order) {
>>? File file(fname, "r", 0);
>>? assert(file);
>
>? ngram = new Ngram(vocab, order);
>>? ngram->read(file, false);
>>? cerr << "Done\n";
>>}
>>
>>
>>int SrilmTest::srilm_getvoc(const char* word) {
>>? return vocab.getIndex((VocabString)word);
>>}
>>
>>
>>float SrilmTest::srilm_wordprob(int w, int* context) {
>>? return (float)ngram->wordProb(w, (VocabIndex*)context);
>>}
>>
>>
>>
>>On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>>
>>Hey,? 
>>>
>>>
>>>I need to use srilm in my c++ code to build ML. all the examples and slides in the internet explained it using ngram-count command not code. I know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>>>
>>>
>>>cheers,
>>>
>>>
>>>
-- 
>>>Mohsen
>>>
>>>_______________________________________________
>>>SRILM-User site list
>>>SRILM-User at speech.sri.com
>>>http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>
>>
>>
>>
>>
-- 
>>Sincerely,
>>Yi Yang
>>http://www.cc.gatech.edu/%7Eyyang319/
>>
>>
>>
>>
>>_______________________________________________
SRILM-User site list SRILM-User at speech.sri.com http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>_______________________________________________
>SRILM-User site list
>SRILM-User at speech.sri.com
>http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>_______________________________________________
>SRILM-User site list
>SRILM-User at speech.sri.com
>http://www.speech.sri.com/mailman/listinfo/srilm-user
>


-- 
Mohsen Jadidi

_______________________________________________
SRILM-User site list
SRILM-User at speech.sri.com
http://www.speech.sri.com/mailman/listinfo/srilm-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130309/b5f38ee3/attachment.html>

From venkataraman.anand at gmail.com  Sat Mar  9 14:50:42 2013
From: venkataraman.anand at gmail.com (Anand Venkataraman)
Date: Sat, 9 Mar 2013 14:50:42 -0800
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <1362868403.15307.YahooMailNeo@web126204.mail.ne1.yahoo.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
	<1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
	<CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
	<1362868403.15307.YahooMailNeo@web126204.mail.ne1.yahoo.com>
Message-ID: <CAF6FMTUfYiPQbwY_NZwfPMHpGQZzzuNV46Zshc8KziuLRRtZxQ@mail.gmail.com>

Solomon

I'm not 100% sure exactly what you're trying to do, but with most
pronunciation modeling, you should be able to get what you want by
inserting the alternate pronunciations in the ASR dictionary. They will be
incorporated as multiple paths within each word in the ngram PFSG prior to
decoding.

&

On Sat, Mar 9, 2013 at 2:33 PM, Solomon Getachew <sgetachew92 at yahoo.com>wrote:

> Dear All
>  I will like developed Multiple pronunciation Modeling in ASR for Amharic
> language based on Knowledge Based i need to get sample code is there any
> one help me?
> thanks in advance
>
>    ------------------------------
> *From:* mohsen jadidi <mohsen.jadidi at gmail.com>
> *To:* yasser hifny <yhifny at yahoo.com>
> *Cc:* "SRILM-User at speech.sri.com" <SRILM-User at speech.sri.com>
> *Sent:* Friday, March 8, 2013 1:10 PM
>
> *Subject:* Re: [SRILM User List] c++ sample for building language model
>
> Thanks for your reply. Very helpful. But don't we need to use ssIndex and
> seIndex to build lexicon language model  and then use it for next step?
>
>
>
>
> On Fri, Mar 8, 2013 at 9:49 PM, yasser hifny <yhifny at yahoo.com> wrote:
>
> Hi Mohsen,
>
> code sample  that I use in my work is
>
> // gloable variables
> Vocab* g_srilm_vocab;
> Ngram* g_model;
>
> double GetLogWordProb(const std::string& strCurrWord,const
> vector<std::string>& vstrHistory)
> {
> float fResult;
>
>  size_t len = 1 + vstrHistory.size()+1;
> VocabIndex* WordIDs=new VocabIndex[len];
>  WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()),
> g_srilm_vocab->unkIndex());
> for(size_t i=0;i<vstrHistory.size();i++)
> WordIDs[i+1] = vstrHistory[i]!="<s>"?
> g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()),
> g_srilm_vocab->unkIndex()):Vocab_None;
>  WordIDs[vstrHistory.size()+1] = Vocab_None;
>
>
>  for(size_t k=0; k<len;k++)
> {
> DEBUG("k=%d wordindex:%d
> wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
> }
> fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
>  if(fResult==LogP_Zero)fResult=0.0;
> DEBUG("prob=%f",fResult);
>
> //g_model->sentenceProb(words, stats);
> delete[] WordIDs;
>
> return fResult;
>
> }
>
>
> //in the  main function
>
>
> //--------------------------------
>  // Load LM
> //--------------------------------
>  g_srilm_vocab= new Vocab;
> g_model=new Ngram(*g_srilm_vocab,nOrder);
>  File file(strLangModelFile.c_str(), "r");
>  if(!file)
> {
>  ERROR("Could not open file %s",strLangModelFile.c_str());
>  }
> g_model->read(file, 0);
>  for(size_t i=0;i<nOrder;i++)
> TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));
>
>
> Best regards,
> Yasser
>
>   ------------------------------
> *From:* Andreas Stolcke <stolcke at icsi.berkeley.edu>
> *To:* Yi Yang <yangyiycc at gmail.com>
> *Cc:* SRILM-User at speech.sri.com
> *Sent:* Friday, March 8, 2013 10:37 PM
> *Subject:* Re: [SRILM User List] c++ sample for building language model
>
>  On 3/8/2013 8:06 AM, Yi Yang wrote:
>
> Hi Mohsen,
>
>  Hope the following codes can be helpful:
>
> you forgot to create the Vocab object.  In your case you could create it
> globally to your code works otherwise:
>
> Vocab vocab;
>
> Andreas
>
>
>  void SrilmTest::srilm_init(const char* fname, int order) {
>   File file(fname, "r", 0);
>   assert(file);
>
>
>     ngram = new Ngram(vocab, order);
>   ngram->read(file, false);
>   cerr << "Done\n";
> }
>
>  int SrilmTest::srilm_getvoc(const char* word) {
>   return vocab.getIndex((VocabString)word);
> }
>
>  float SrilmTest::srilm_wordprob(int w, int* context) {
>   return (float)ngram->wordProb(w, (VocabIndex*)context);
>  }
>
>
> On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>
> Hey,
>
>  I need to use srilm in my c++ code to build ML. all the examples and
> slides in the internet explained it using ngram-count command not code. I
> know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>
>  cheers,
>
>
>  --
> Mohsen
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
>
>  --
> Sincerely,
> Yi Yang
> http://www.cc.gatech.edu/%7Eyyang319/
>
>
> _______________________________________________
> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>
>
> --
> Mohsen Jadidi
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130309/5c7e3d36/attachment.html>

From mohsen.jadidi at gmail.com  Sun Mar 10 06:02:45 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Sun, 10 Mar 2013 14:02:45 +0100
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <CAF6FMTUfYiPQbwY_NZwfPMHpGQZzzuNV46Zshc8KziuLRRtZxQ@mail.gmail.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
	<1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
	<CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
	<1362868403.15307.YahooMailNeo@web126204.mail.ne1.yahoo.com>
	<CAF6FMTUfYiPQbwY_NZwfPMHpGQZzzuNV46Zshc8KziuLRRtZxQ@mail.gmail.com>
Message-ID: <CAOfo52Bpf01KJwRXwxnCqOdNH19_+t4AkwbT0DsNpJ4HQRRM9A@mail.gmail.com>

hi solomon,

I think it was better to open a new topic for this problem


On Sat, Mar 9, 2013 at 11:50 PM, Anand Venkataraman <
venkataraman.anand at gmail.com> wrote:

> Solomon
>
> I'm not 100% sure exactly what you're trying to do, but with most
> pronunciation modeling, you should be able to get what you want by
> inserting the alternate pronunciations in the ASR dictionary. They will be
> incorporated as multiple paths within each word in the ngram PFSG prior to
> decoding.
>
> &
>
> On Sat, Mar 9, 2013 at 2:33 PM, Solomon Getachew <sgetachew92 at yahoo.com>wrote:
>
>> Dear All
>>  I will like developed Multiple pronunciation Modeling in ASR for Amharic
>> language based on Knowledge Based i need to get sample code is there any
>> one help me?
>> thanks in advance
>>
>>    ------------------------------
>> *From:* mohsen jadidi <mohsen.jadidi at gmail.com>
>> *To:* yasser hifny <yhifny at yahoo.com>
>> *Cc:* "SRILM-User at speech.sri.com" <SRILM-User at speech.sri.com>
>> *Sent:* Friday, March 8, 2013 1:10 PM
>>
>> *Subject:* Re: [SRILM User List] c++ sample for building language model
>>
>> Thanks for your reply. Very helpful. But don't we need to use ssIndex and
>> seIndex to build lexicon language model  and then use it for next step?
>>
>>
>>
>>
>> On Fri, Mar 8, 2013 at 9:49 PM, yasser hifny <yhifny at yahoo.com> wrote:
>>
>> Hi Mohsen,
>>
>> code sample  that I use in my work is
>>
>> // gloable variables
>> Vocab* g_srilm_vocab;
>> Ngram* g_model;
>>
>> double GetLogWordProb(const std::string& strCurrWord,const
>> vector<std::string>& vstrHistory)
>> {
>> float fResult;
>>
>>  size_t len = 1 + vstrHistory.size()+1;
>> VocabIndex* WordIDs=new VocabIndex[len];
>>  WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()),
>> g_srilm_vocab->unkIndex());
>> for(size_t i=0;i<vstrHistory.size();i++)
>> WordIDs[i+1] = vstrHistory[i]!="<s>"?
>> g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()),
>> g_srilm_vocab->unkIndex()):Vocab_None;
>>  WordIDs[vstrHistory.size()+1] = Vocab_None;
>>
>>
>>  for(size_t k=0; k<len;k++)
>> {
>> DEBUG("k=%d wordindex:%d
>> wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
>> }
>> fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
>>  if(fResult==LogP_Zero)fResult=0.0;
>> DEBUG("prob=%f",fResult);
>>
>> //g_model->sentenceProb(words, stats);
>> delete[] WordIDs;
>>
>> return fResult;
>>
>> }
>>
>>
>> //in the  main function
>>
>>
>> //--------------------------------
>>  // Load LM
>> //--------------------------------
>>  g_srilm_vocab= new Vocab;
>> g_model=new Ngram(*g_srilm_vocab,nOrder);
>>  File file(strLangModelFile.c_str(), "r");
>>  if(!file)
>> {
>>  ERROR("Could not open file %s",strLangModelFile.c_str());
>>  }
>> g_model->read(file, 0);
>>  for(size_t i=0;i<nOrder;i++)
>> TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));
>>
>>
>> Best regards,
>> Yasser
>>
>>   ------------------------------
>> *From:* Andreas Stolcke <stolcke at icsi.berkeley.edu>
>> *To:* Yi Yang <yangyiycc at gmail.com>
>> *Cc:* SRILM-User at speech.sri.com
>> *Sent:* Friday, March 8, 2013 10:37 PM
>> *Subject:* Re: [SRILM User List] c++ sample for building language model
>>
>>  On 3/8/2013 8:06 AM, Yi Yang wrote:
>>
>> Hi Mohsen,
>>
>>  Hope the following codes can be helpful:
>>
>> you forgot to create the Vocab object.  In your case you could create it
>> globally to your code works otherwise:
>>
>> Vocab vocab;
>>
>> Andreas
>>
>>
>>  void SrilmTest::srilm_init(const char* fname, int order) {
>>   File file(fname, "r", 0);
>>   assert(file);
>>
>>
>>     ngram = new Ngram(vocab, order);
>>   ngram->read(file, false);
>>   cerr << "Done\n";
>> }
>>
>>  int SrilmTest::srilm_getvoc(const char* word) {
>>   return vocab.getIndex((VocabString)word);
>> }
>>
>>  float SrilmTest::srilm_wordprob(int w, int* context) {
>>   return (float)ngram->wordProb(w, (VocabIndex*)context);
>>  }
>>
>>
>> On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>>
>> Hey,
>>
>>  I need to use srilm in my c++ code to build ML. all the examples and
>> slides in the internet explained it using ngram-count command not code. I
>> know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>>
>>  cheers,
>>
>>
>>  --
>> Mohsen
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>>
>>
>>  --
>> Sincerely,
>> Yi Yang
>> http://www.cc.gatech.edu/%7Eyyang319/
>>
>>
>> _______________________________________________
>> SRILM-User site listSRILM-User at speech.sri.comhttp://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>>
>>
>> --
>> Mohsen Jadidi
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>>
>> _______________________________________________
>> SRILM-User site list
>> SRILM-User at speech.sri.com
>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>
>
>
> _______________________________________________
> SRILM-User site list
> SRILM-User at speech.sri.com
> http://www.speech.sri.com/mailman/listinfo/srilm-user
>



-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130310/d6ded38b/attachment.html>

From yhifny at yahoo.com  Sun Mar 10 06:09:20 2013
From: yhifny at yahoo.com (yasser hifny)
Date: Sun, 10 Mar 2013 06:09:20 -0700 (PDT)
Subject: [SRILM User List] c++ sample for building language model
In-Reply-To: <CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
References: <CAOfo52C_=PEnd5JZzv9y1tPCJWOpy3CpwYyUO7sm3E4w6-zS2g@mail.gmail.com>
	<CAKNTL4g7nOWVHNS5P3VgrrqzzsAxk389LH4pcDs5PYziPkUJ=A@mail.gmail.com>
	<513A4C0F.4070104@icsi.berkeley.edu>
	<1362775793.14959.YahooMailNeo@web125806.mail.ne1.yahoo.com>
	<CAOfo52CWkA=7MbWgnMdgaTYtvghYMQKoTR+OEXeT+9DfmrHvpA@mail.gmail.com>
Message-ID: <1362920960.62871.YahooMailNeo@web125804.mail.ne1.yahoo.com>

Dear?Mohsen,
I do not understand your question.?perhaps,?the group?members?can help.

best regards,
Yasser


________________________________
 From: mohsen jadidi <mohsen.jadidi at gmail.com>
To: yasser hifny <yhifny at yahoo.com> 
Cc: "SRILM-User at speech.sri.com" <SRILM-User at speech.sri.com> 
Sent: Friday, March 8, 2013 11:10 PM
Subject: Re: [SRILM User List] c++ sample for building language model
 

Thanks for your reply. Very helpful. But don't we need to use ssIndex and seIndex to build lexicon language model ?and then use it for next step??





On Fri, Mar 8, 2013 at 9:49 PM, yasser hifny <yhifny at yahoo.com> wrote:

Hi Mohsen,
>
>
>code sample ?that I use in my work is
>
>
>// gloable variables
>Vocab*g_srilm_vocab;
>
>Ngram*g_model;
>
>
>double GetLogWordProb(const std::string& strCurrWord,const vector<std::string>& vstrHistory)
>{
>floatfResult;
>
>
>size_t len = 1 + vstrHistory.size()+1;
>VocabIndex* WordIDs=new VocabIndex[len];
>WordIDs[0] = g_srilm_vocab->getIndex((char*)(strCurrWord.c_str()), g_srilm_vocab->unkIndex());
>for(size_t i=0;i<vstrHistory.size();i++)
>WordIDs[i+1] = vstrHistory[i]!="<s>"?
>g_srilm_vocab->getIndex((char*)(vstrHistory[i].c_str()), g_srilm_vocab->unkIndex()):Vocab_None;
>WordIDs[vstrHistory.size()+1] = Vocab_None;
>
>
>
>
>for(size_t k=0; k<len;k++)
>{
>DEBUG("k=%d wordindex:%d wordstring:%s",k,WordIDs[k],g_srilm_vocab->getWord(WordIDs[k]));
>}
>fResult=g_model->wordProb(WordIDs[0], &WordIDs[1]);
>if(fResult==LogP_Zero)fResult=0.0;
>DEBUG("prob=%f",fResult);
>
>
>//g_model->sentenceProb(words, stats);
>delete[] WordIDs;
>
>
>return fResult;
>
>
>}
>
>
>
>
>//in the ?main function
>
>
>
>
>//--------------------------------
>// Load LM
>//--------------------------------
>g_srilm_vocab= new Vocab;
>g_model=new Ngram(*g_srilm_vocab,nOrder);
>File file(strLangModelFile.c_str(), "r");
>if(!file)
>{
>ERROR("Could not open file %s",strLangModelFile.c_str());
>}
>g_model->read(file, 0);
>for(size_t i=0;i<nOrder;i++)
>TRACE("Num of ngram in model order %d:%d",i+1,g_model->numNgrams(i+1));
>
>
>
>
>Best regards,
>Yasser
>
>
>
>________________________________
> From: Andreas Stolcke <stolcke at icsi.berkeley.edu>
>To: Yi Yang <yangyiycc at gmail.com> 
>Cc: SRILM-User at speech.sri.com 
>Sent: Friday, March 8, 2013 10:37 PM
>Subject: Re: [SRILM User List] c++ sample for building language model
> 
>
>On 3/8/2013 8:06 AM, Yi Yang wrote:
>
>Hi Mohsen, 
>>
>>
>>Hope the following codes can be helpful:
you forgot to create the Vocab object.? In your case you could create it globally to your code works otherwise:
>
>Vocab vocab;
>
>Andreas
>
>
>
>>
>>void SrilmTest::srilm_init(const char* fname, int order) {
>>? File file(fname, "r", 0);
>>? assert(file);
>
>? ngram = new Ngram(vocab, order);
>>? ngram->read(file, false);
>>? cerr << "Done\n";
>>}
>>
>>
>>int SrilmTest::srilm_getvoc(const char* word) {
>>? return vocab.getIndex((VocabString)word);
>>}
>>
>>
>>float SrilmTest::srilm_wordprob(int w, int* context) {
>>? return (float)ngram->wordProb(w, (VocabIndex*)context);
>>}
>>
>>
>>
>>On Thu, Mar 7, 2013 at 3:23 PM, mohsen jadidi <mohsen.jadidi at gmail.com> wrote:
>>
>>Hey,? 
>>>
>>>
>>>I need to use srilm in my c++ code to build ML. all the examples and slides in the internet explained it using ngram-count command not code. I know should use <Ngram.h> <Vocab.h>. can you point me to a starting point?
>>>
>>>
>>>cheers,
>>>
>>>
>>>
-- 
>>>Mohsen
>>>
>>>_______________________________________________
>>>SRILM-User site list
>>>SRILM-User at speech.sri.com
>>>http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>
>>
>>
>>
>>
-- 
>>Sincerely,
>>Yi Yang
>>http://www.cc.gatech.edu/%7Eyyang319/
>>
>>
>>
>>
>>_______________________________________________
SRILM-User site list SRILM-User at speech.sri.com http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>_______________________________________________
>SRILM-User site list
>SRILM-User at speech.sri.com
>http://www.speech.sri.com/mailman/listinfo/srilm-user
>
>
>_______________________________________________
>SRILM-User site list
>SRILM-User at speech.sri.com
>http://www.speech.sri.com/mailman/listinfo/srilm-user
>


-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130310/fe30f194/attachment.html>

From dcherubangel at gmail.com  Mon Mar 11 07:09:44 2013
From: dcherubangel at gmail.com (Nutthamon)
Date: Mon, 11 Mar 2013 21:09:44 +0700
Subject: [SRILM User List] how to genereate character lm model
Message-ID: <CANFpwQ2En5swTVjaF9yk2FmtrLFS+7zK=pX6_N5dGxsqdiURMQ@mail.gmail.com>

Hello,

I am new to language modeling and SRILM toolkit.

Is this toolkit can generate language model in character level? If can
do that, what is a command for do that i can't find it.And please give
example to me.


i used this tool via cygwin terminal.

Example in training.txt

s i m p l y
g o o d
t h a n k y o u
c l o u n d

or
training2.txt
s i m p l y g o o d t h a n k y o u c l o u n d


which training text correct for LM built on character level?
first,right? If first i can directly enter to add more line or add
some symbol for add line?

i'm not sure what is <s>and</s> mean.

Is this command for train lm model character level (trigram)?
ngram-count -order 4 -text /srilm/sences.txt -write /srilm/corpus

i try training1.txt resul is

<s>     1
<s> ??S 1
<s> ??S </s>    1
??S     1
??S </s>        1
</s>
i don't know what is it wht it's not count character but when i try
word level. the result is sum of count of word

Many thank in advance

-- 
Best Regards,
Nutthamon Moknarong
dcherubangel at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130311/f2151df1/attachment.html>

From mohsen.jadidi at gmail.com  Wed Mar 13 08:03:20 2013
From: mohsen.jadidi at gmail.com (mohsen jadidi)
Date: Wed, 13 Mar 2013 16:03:20 +0100
Subject: [SRILM User List] installation on mac lion error on negram class
In-Reply-To: <CAAbXKqfjudAXmYqii-As8Z1NV9jX8gYpN=cPsLwyZduTpc9cPw@mail.gmail.com>
References: <CAOfo52BqrcSQ8qUAbmD0TqVrAq0krzOruK+7AH0_Nw9xnuF8TQ@mail.gmail.com>
	<51341587.6000001@icsi.berkeley.edu>
	<CAOfo52C0QM79EC+CQk_oc+YtTAtg5UX7H0cr6JT04moANQH9gg@mail.gmail.com>
	<7874D710-02DF-48B2-90D4-8E7E858799D5@gmail.com>
	<CAOfo52Ajekw0Ji9myik=9p+0m_SyFjHwo-aVR4v+xWjY3VLF+A@mail.gmail.com>
	<CAAbXKqcx4vqYR7NQVxEELOUYb9JrSQHsZ5VjUFvjiz_QSZP_cg@mail.gmail.com>
	<CAOfo52AQCkuvV313j7WJdv3w0UvqM+yESQH-FqMfV3r6M+N2AQ@mail.gmail.com>
	<CAAbXKqfnw6=bx8oKiuOri4UCaqk7DrrDtm38Qb_CE=xxnNN0MA@mail.gmail.com>
	<CAOfo52CyHQawme-4zjpCph41sfv6Rjfg6t2CHi=5DuyUS2w4iA@mail.gmail.com>
	<CAAbXKqfRbX60o7hXnmR_n_NHUePiEcFOX-3TRVr_HvBYPO7wXQ@mail.gmail.com>
	<CAOfo52BOLPswY-3SrrPgKix=ONCOnu+mjw+deo=Vp=D1h--uag@mail.gmail.com>
	<CAAbXKqcBTHFCaCw1evLO=-pFe+6brwOr3Z65M1bNyg=atYAADA@mail.gmail.com>
	<CAOfo52B9mSQ4K-HQhuCogFC-9MwbEV6yWGw5E6RDFRgdqZ_o8g@mail.gmail.com>
	<CAAbXKqeuHvxZLvdyfKcdbYPLXTZLGfH+h-85==83FTgnh9VFow@mail.gmail.com>
	<CAOfo52B8odWjfb=HGJUD3FSn2vNoDvrQd9rOjo=D9=3+pA6G9w@mail.gmail.com>
	<CAOfo52DxNMRZ8eECVbk_pyW8qdkCwMcW0PKP-VN_4Skm7bxxyA@mail.gmail.com>
	<CAAbXKqf129LOYrSQxdUtGacis_91=zYJ5_NQn8=4g=YBb32-5A@mail.gmail.com>
	<CAOfo52AUJ+UA4eRrnGTRBckGPg97YRSQ-a8pqSHeP-kRM-w5zg@mail.gmail.com>
	<CAAbXKqfjudAXmYqii-As8Z1NV9jX8gYpN=cPsLwyZduTpc9cPw@mail.gmail.com>
Message-ID: <CAOfo52ARnG9+A7Y0Zv6htTtYP7yYMEbbC9Z-=aHebSariNUfCg@mail.gmail.com>

I managed to run my program after installing the srilm
with NO_ICONV=anything . Otherwise it keeps giving me this error:


Undefined symbols for architecture x86_64:
  "_iconv", referenced from:
      File::fgetsUTF8(char*, int)in libmisc.a(File.o)
  "_iconv_close", referenced from:
      File::reopen(char const*)in libmisc.a(File.o)
      File::reopen(std::basic_string<char, std::char_traits<char>,
std::allocator<char> >&, int)in libmisc.a(File.o)
      File::reopen(char const*, unsigned long, int)in libmisc.a(File.o)
      File::reopen(char const*, char const*)in libmisc.a(File.o)
      File::~File() in libmisc.a(File.o)
      File::~File() in libmisc.a(File.o)
  "_iconv_open", referenced from:
      File::fgetsUTF8(char*, int)in libmisc.a(File.o)
ld: symbol(s) not found for architecture x86_64
collect2: ld returned 1 exit status

I read some pages and they suggested problem comes from the different
version of iconv,(The Mac default library and Macport installation).  I
couldn't realize how to fix the problem. I tried to set   ADDITIONAL_CFLAGS
= /usr/lib/libiconv.2.dylib but didnt work.
These are some information you might find useful :
For default version in /usr/lib I have :

ls -ll libiconv*
lrwxr-xr-x  1 root  wheel       16 Oct  8  2011 libiconv.2.4.0.dylib ->
libiconv.2.dylib
-r-xr-xr-x  1 root  wheel  2105216 Oct  8  2011 libiconv.2.dylib
lrwxr-xr-x  1 root  wheel       20 Oct  8  2011 libiconv.dylib ->
libiconv.2.4.0.dylib

none of them have (*) to indicate which one my compiler use?! Also,

nm libiconv.2.dylib | grep iconv
00000000000f1af0 S ___iconv_2VersionNumber
00000000000f1b90 S ___iconv_2VersionString
00000000000f60f0 D __libiconv_version
000000000000a1e1 T _iconv
000000000000a5a0 T _iconv_canonicalize
0000000000013164 T _iconv_close
0000000000013171 T _iconv_open
000000000000a72c T _iconvctl
000000000000a20f T _iconvlist
0000000000014dbd T _libiconv_relocate
0000000000014cff T _libiconv_set_relocation_prefix

For Macport version I have :

-rw-r--r--  1 root  admin  1072264 Apr  4  2012 libiconv.2.dylib
-rw-r--r--  1 root  admin  1098856 Apr  4  2012 libiconv.a
lrwxr-xr-x  1 root  admin       16 Apr  4  2012 libiconv.dylib ->
libiconv.2.dylib
-rw-r--r--  1 root  admin      914 Apr  4  2012 libiconv.la

and also:

 nm libiconv.a | grep iconv
libiconv.a(iconv.o):
0000000000016780 D __libiconv_version
000000000000ac10 T _iconv_canonicalize
00000000000f9908 S _iconv_canonicalize.eh
000000000000a810 T _libiconv
00000000000f97d0 S _libiconv.eh
00000000000159f0 T _libiconv_close
00000000000fa6c0 S _libiconv_close.eh
0000000000015a00 T _libiconv_open
00000000000fa6f0 S _libiconv_open.eh
0000000000014950 T _libiconv_open_into
00000000000fa518 S _libiconv_open_into.eh
000000000000adc0 T _libiconvctl
00000000000f9940 S _libiconvctl.eh
000000000000a850 T _libiconvlist
00000000000f9830 S _libiconvlist.eh
libiconv.a(localcharset.o):
libiconv.a(relocatable.o):
00000000000000c0 T _libiconv_relocate
00000000000001d0 S _libiconv_relocate.eh
0000000000000000 T _libiconv_set_relocation_prefix
0000000000000198 S _libiconv_set_relocation_prefix.eh


Do you have any suggestion ?


On Tue, Mar 12, 2013 at 10:53 PM, Charles Wooters <wooters at icsi.berkeley.edu
> wrote:

> That is weird! So, I'm running 10.8.2 with Xcode 4.6, is that what you
> have too? And did you try a fresh install?
>
> If the answer to those two questions are "yes" and "yes", then could you
> send me a small example program or command line that causes the problem?
> Maybe my install is faulty too and I just haven't run into it yet...
>
> ---
> Chuck Wooters
> International Computer Science Institute
> http://www.icsi.berkeley.edu
>
>
> On Tue, Mar 12, 2013 at 2:41 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>
>> c++ gives me the same error!
>>
>>
>> On Mon, Mar 11, 2013 at 5:49 PM, Charles Wooters <
>> wooters at icsi.berkeley.edu> wrote:
>>
>>> Hi Mohsen,
>>>
>>> I'm not sure what is happening. I have only ever compiled successfully
>>> using the LLVM-based compiler - 'c++'. I have never tried with 'g++'. I
>>> suspect that there is some missing or mismatched library at runtime. Have
>>> you tried using 'c++' yet?
>>>
>>> -Chuck
>>>
>>>
>>> ---
>>> Chuck Wooters
>>> International Computer Science Institute
>>> http://www.icsi.berkeley.edu
>>>
>>>
>>> On Sun, Mar 10, 2013 at 2:47 PM, mohsen jadidi <mohsen.jadidi at gmail.com>wrote:
>>>
>>>> I compile and link like this:
>>>>
>>>>  g++ -c -o lm.o LanguageModel.cpp
>>>> -I/Users/mohsenjadidi/Documents/srilm/include  #it is ok
>>>> g++ -o lm lm.o -L/Users/mohsenjadidi/Documents/srilm/lib/macosx-m64
>>>>  #(ERROR)
>>>>
>>>>
>>>>
>>>>
>>>> On Sun, Mar 10, 2013 at 10:45 PM, mohsen jadidi <
>>>> mohsen.jadidi at gmail.com> wrote:
>>>>
>>>>> Hello Charles,
>>>>>
>>>>> I stil have problem with this library. When ever instantiate Vocab or
>>>>> Ngam class ( vocal_srilm = new Vocab) I am getting this error :
>>>>>
>>>>> Undefined symbols for architecture x86_64:
>>>>>   "Ngram::Ngram(Vocab&, unsigned int)", referenced from:
>>>>>       _main in lm.o
>>>>> ld: symbol(s) not found for architecture x86_64
>>>>> collect2: ld returned 1 exit status
>>>>>
>>>>> Do you have any idea?
>>>>>
>>>>>
>>>>>
>>>>> On Wed, Mar 6, 2013 at 6:29 PM, Charles Wooters <
>>>>> wooters at icsi.berkeley.edu> wrote:
>>>>>
>>>>>> Great!
>>>>>>
>>>>>> By the way, if you are interested in building maximum entropy models
>>>>>> using libLBFGS optimization library (as described in the 'INSTALL' file), I
>>>>>> have written up some instructions about how to compile libLBFGS on the Mac
>>>>>> and hook it into the SRILM Toolkit. You can find the instructions here:
>>>>>>
>>>>>> https://gist.github.com/wooters/5095712
>>>>>>
>>>>>> -Chuck
>>>>>>
>>>>>> ---
>>>>>> Chuck Wooters
>>>>>> International Computer Science Institute
>>>>>> http://www.icsi.berkeley.edu
>>>>>>
>>>>>>
>>>>>> On Wed, Mar 6, 2013 at 1:04 AM, mohsen jadidi <
>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>
>>>>>>> yay..worked.the problem was the gcc version. I set the default one
>>>>>>> and it worked.Thank you all
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Mar 6, 2013 at 12:11 AM, Charles Wooters <
>>>>>>> wooters at icsi.berkeley.edu> wrote:
>>>>>>>
>>>>>>>> I am just using the default compiler that comes with Xcode - I
>>>>>>>> never installed gcc through homebrew. So, maybe try that...
>>>>>>>>
>>>>>>>> ---
>>>>>>>> Chuck Wooters
>>>>>>>> International Computer Science Institute
>>>>>>>> http://www.icsi.berkeley.edu
>>>>>>>>
>>>>>>>>
>>>>>>>> On Tue, Mar 5, 2013 at 3:00 PM, mohsen jadidi <
>>>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>>>
>>>>>>>>> the thing is I updated my gcc to 4.7 using homebrew and set it as
>>>>>>>>> my default. Is it a problem?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Tue, Mar 5, 2013 at 11:56 PM, Charles Wooters <
>>>>>>>>> wooters at icsi.berkeley.edu> wrote:
>>>>>>>>>
>>>>>>>>>> Hmmm. It looks like there may be some sort of compiler mismatch
>>>>>>>>>> here (see this article:
>>>>>>>>>> http://stackoverflow.com/questions/4035445/linker-error-undefined-reference-to-stdctypechar-m-widen-init
>>>>>>>>>> )
>>>>>>>>>>
>>>>>>>>>> If you haven't already, I suggest that you start over:
>>>>>>>>>> re-download the toolkit, untar it into a new directory, and then compile.
>>>>>>>>>> See if that fixes things.
>>>>>>>>>>
>>>>>>>>>> -Chuck
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> ---
>>>>>>>>>> Chuck Wooters
>>>>>>>>>> International Computer Science Institute
>>>>>>>>>> http://www.icsi.berkeley.edu
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On Tue, Mar 5, 2013 at 2:47 PM, mohsen jadidi <
>>>>>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>>>>>
>>>>>>>>>>> I installed those libraries and update my xcode but didnt fix :
>>>>>>>>>>>
>>>>>>>>>>> or subdir in misc dstruct lm flm lattice utils; do \
>>>>>>>>>>> (cd $subdir/src; make SRILM=/Users/mohsenjadidi/Downloads/srilm
>>>>>>>>>>> MACHINE_TYPE=macosx-m64 OPTION= MAKE_PIC= release-libraries) || exit 1; \
>>>>>>>>>>>  done
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-libraries'.
>>>>>>>>>>> make release-programs
>>>>>>>>>>> for subdir in misc dstruct lm flm lattice utils; do \
>>>>>>>>>>> (cd $subdir/src; make SRILM=/Users/mohsenjadidi/Downloads/srilm
>>>>>>>>>>> MACHINE_TYPE=macosx-m64 OPTION= MAKE_PIC= release-programs) || exit 1; \
>>>>>>>>>>>  done
>>>>>>>>>>> make[2]: Nothing to be done for `release-programs'.
>>>>>>>>>>> make[2]: Nothing to be done for `release-programs'.
>>>>>>>>>>> c++ -Wreturn-type -Wimplicit -m64 -DINSTANTIATE_TEMPLATES
>>>>>>>>>>> -I/usr/include -I. -I../../include -DHAVE_ZOPEN   -u _matherr
>>>>>>>>>>> -L../../lib/macosx-m64  -g -O2 -fno-common -o ../bin/macosx-m64/ngram
>>>>>>>>>>> ../obj/macosx-m64/ngram.o ../obj/macosx-m64/liboolm.a
>>>>>>>>>>>  ../../lib/macosx-m64/libflm.a ../../lib/macosx-m64/libdstruct.a
>>>>>>>>>>> ../../lib/macosx-m64/libmisc.a -L/usr/lib -ltcl -lm   2>&1 | c++filt
>>>>>>>>>>>  Undefined symbols for architecture x86_64:
>>>>>>>>>>>   "std::ctype<char>::_M_widen_init() const", referenced from:
>>>>>>>>>>>       makeLoglinearMixLM(Array<char const*>, Vocab&, SubVocab*,
>>>>>>>>>>> unsigned int, LM*, Array<double>) in ngram.o
>>>>>>>>>>>       LHash<unsigned int, unsigned int*>::dump() const in
>>>>>>>>>>> liboolm.a(MultiwordVocab.o)
>>>>>>>>>>>       LHash<char const*, unsigned int*>::dump() const in
>>>>>>>>>>> liboolm.a(RefList.o)
>>>>>>>>>>>       ClassNgram::prefixProb(unsigned int, unsigned int const*,
>>>>>>>>>>> float&, TextStats&) in liboolm.a(ClassNgram.o)
>>>>>>>>>>>       ClassNgram::expand(unsigned int, unsigned int) in
>>>>>>>>>>> liboolm.a(ClassNgram.o)
>>>>>>>>>>>       NgramCounts<float>::readBinaryNode(Trie<unsigned int,
>>>>>>>>>>> float>&, unsigned int, unsigned int, File&, long long&, bool,
>>>>>>>>>>> Array<unsigned int>&) in liboolm.a(ClassNgram.o)
>>>>>>>>>>>       NgramCounts<float>::readBinary(File&, unsigned int, bool)
>>>>>>>>>>> in liboolm.a(ClassNgram.o)
>>>>>>>>>>>       ...
>>>>>>>>>>> ld: symbol(s) not found for architecture x86_64
>>>>>>>>>>> collect2: error: ld returned 1 exit status
>>>>>>>>>>> test -f ../bin/macosx-m64/ngram
>>>>>>>>>>> make[2]: *** [../bin/macosx-m64/ngram] Error 1
>>>>>>>>>>> make[1]: *** [release-programs] Error 1
>>>>>>>>>>> make: *** [World] Error 2
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On Tue, Mar 5, 2013 at 5:52 PM, Charles Wooters <
>>>>>>>>>>> wooters at icsi.berkeley.edu> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> As for prerequisites, I installed these two using the homebrew
>>>>>>>>>>>> package manager (http://mxcl.github.com/homebrew/):
>>>>>>>>>>>>
>>>>>>>>>>>> - gawk (this also installed xz)
>>>>>>>>>>>> - gzip
>>>>>>>>>>>>
>>>>>>>>>>>> These were already installed on my machine (perhaps by Xcode or
>>>>>>>>>>>> Xcode command-line utilities):
>>>>>>>>>>>>
>>>>>>>>>>>>  - make
>>>>>>>>>>>>  - tcsh
>>>>>>>>>>>>  - tcl
>>>>>>>>>>>>  - bzip2
>>>>>>>>>>>>
>>>>>>>>>>>> This is the only one that I haven't installed (but I will now):
>>>>>>>>>>>>
>>>>>>>>>>>>  - p7zip
>>>>>>>>>>>>
>>>>>>>>>>>> Hope that helps.
>>>>>>>>>>>>
>>>>>>>>>>>> -Chuck
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> ---
>>>>>>>>>>>> Chuck Wooters
>>>>>>>>>>>> International Computer Science Institute
>>>>>>>>>>>> http://www.icsi.berkeley.edu
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> On Mon, Mar 4, 2013 at 10:46 PM, mohsen jadidi <
>>>>>>>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>> did you install any of prerequisites libraries?
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> On Tue, Mar 5, 2013 at 12:26 AM, Charles Wooters <
>>>>>>>>>>>>> wooters at icsi.berkeley.edu> wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>> I was able to compile v1.7 on my macbook pro (2.6 GHz Intel
>>>>>>>>>>>>>> Core i7) running OSX 10.8.2 following these instructions:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> http://www1.icsi.berkeley.edu/~wooters/SRILM/index.html
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> -Chuck
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> ---
>>>>>>>>>>>>>> Chuck Wooters
>>>>>>>>>>>>>> International Computer Science Institute
>>>>>>>>>>>>>> http://www.icsi.berkeley.edu
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> On Mon, Mar 4, 2013 at 12:35 PM, mohsen jadidi <
>>>>>>>>>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> this is mine:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2
>>>>>>>>>>>>>>> architectures
>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O
>>>>>>>>>>>>>>> 64-bit dynamically linked shared library x86_64
>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O
>>>>>>>>>>>>>>> dynamically linked shared library i386
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> On Mon, Mar 4, 2013 at 8:20 PM, Roman Kutlak <
>>>>>>>>>>>>>>> kutlak.roman at gmail.com> wrote:
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> You should check that you have libiconv in /usr/lib and
>>>>>>>>>>>>>>>> that you have the 64bit version using the following command:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> file /usr/lib/libiconv.dylib
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>  You should see something like the following:
>>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib: Mach-O universal binary with 2
>>>>>>>>>>>>>>>> architectures
>>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib (for architecture i386): Mach-O
>>>>>>>>>>>>>>>> dynamically linked shared library i386
>>>>>>>>>>>>>>>> /usr/lib/libiconv.dylib (for architecture x86_64): Mach-O
>>>>>>>>>>>>>>>> 64-bit dynamically linked shared library x86_64
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> Roman
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> On 4 Mar 2013, at 18:40, mohsen jadidi <
>>>>>>>>>>>>>>>> mohsen.jadidi at gmail.com> wrote:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> Thanks but it didnt fixed the problem! Just removed the
>>>>>>>>>>>>>>>> warnings
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> On Mon, Mar 4, 2013 at 4:31 AM, Andreas Stolcke <
>>>>>>>>>>>>>>>> stolcke at icsi.berkeley.edu> wrote:
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> On 3/3/2013 1:46 PM, mohsen jadidi wrote:
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> Hello everybody,
>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>> I am trying to install versiobn 1.7 on mac, but it keeps
>>>>>>>>>>>>>>>>>> giving me error:
>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> The cc1plus warning should disappear if you change
>>>>>>>>>>>>>>>>> common/Makefile.machine.macosx  to have
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>    GCC_FLAGS = -Wreturn-type
>>>>>>>>>>>>>>>>>    CC = cc $(GCC_FLAGS) -Wimplicit -Wimplicit-int
>>>>>>>>>>>>>>>>>    CXX = c++ $(GCC_FLAGS) -DINSTANTIATE_TEMPLATES
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> The link error seems to have something to do with the
>>>>>>>>>>>>>>>>> iconv library not being properly installed for 64bit builds.
>>>>>>>>>>>>>>>>> You might be able to work around this by building with
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>     make NO_ICONV=1  ...
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> I hope someone with MacOSX savvy can provide a way to make
>>>>>>>>>>>>>>>>> the linking with -liconv actually work.
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>> Andreas
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> --
>>>>>>>>>>>>>>>> Mohsen Jadidi
>>>>>>>>>>>>>>>>  _______________________________________________
>>>>>>>>>>>>>>>> SRILM-User site list
>>>>>>>>>>>>>>>> SRILM-User at speech.sri.com
>>>>>>>>>>>>>>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>>> SRILM-User site list
>>>>>>>>>>>>>>>> SRILM-User at speech.sri.com
>>>>>>>>>>>>>>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> --
>>>>>>>>>>>>>>> Mohsen Jadidi
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>> SRILM-User site list
>>>>>>>>>>>>>>> SRILM-User at speech.sri.com
>>>>>>>>>>>>>>> http://www.speech.sri.com/mailman/listinfo/srilm-user
>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> --
>>>>>>>>>>>>> Mohsen Jadidi
>>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> Mohsen Jadidi
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Mohsen Jadidi
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Mohsen Jadidi
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Mohsen Jadidi
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Mohsen Jadidi
>>>>
>>>
>>>
>>
>>
>> --
>> Mohsen Jadidi
>>
>
>


-- 
Mohsen Jadidi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130313/31cd62a9/attachment.html>

From shubhamay.sen at gmail.com  Wed Mar 20 23:53:51 2013
From: shubhamay.sen at gmail.com (SHUBHAMAY SEN)
Date: Thu, 21 Mar 2013 12:23:51 +0530
Subject: [SRILM User List] Fwd: Problem- "cannot execute binary file"
In-Reply-To: <CAH7UGApf6ssdVez+wxXwdPLhbgHMj4ifc4T5RRJ3gY74skLBhA@mail.gmail.com>
References: <CAH7UGApf6ssdVez+wxXwdPLhbgHMj4ifc4T5RRJ3gY74skLBhA@mail.gmail.com>
Message-ID: <CAH7UGAo4mJ7nJtQi-GgiOSjdmdhKe68dALNC0h7eQkNhQ6mrjA@mail.gmail.com>

---------- Forwarded message ----------
From: SHUBHAMAY SEN <shubhamay.sen at gmail.com>
Date: Thu, Mar 21, 2013 at 12:14 PM
Subject: Problem- "cannot execute binary file"
To: srilm-user at speech.sri.com


Hi,
      I have a 64 bit machine.

kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$ uname -a
Linux kiit-ThinkCentre-M72e 3.5.0-21-generic #32-Ubuntu SMP Tue Dec 11
18:52:46 UTC 2012 i686 i686 i686 GNU/Linux

I have compiled srilm using "make World MAKE_PIC=1 MACHINE_TYPE=i686-m64 -C
$SRILM" command, output is attached.

but when I am executing the below command, it is showing the following
message.

kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$ $SRILM/bin/*/ngram-count
-order 3
-interpolate -kndiscount -unk -text data/training/news-commentary.tok.en
-lm lm/news.lm.gz

bash: /home/kiit/srilm/bin/i686-m64/ngram-count: cannot execute binary file

Please Suggest...

-- 
---------------------------------------------------------------------
Regards:
Shubhamay Sen
M. Tech CSE
KIIT University, BBSR,
Odisha, India
mob- 09778772782



-- 
---------------------------------------------------------------------
Regards:
Shubhamay Sen
M. Tech CSE
KIIT University, BBSR,
Odisha, India
mob- 09778772782
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130321/e9c09ab5/attachment.html>
-------------- next part --------------
make: Entering directory `/home/kiit/srilm'
mkdir -p include lib bin
make init
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 init) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/misc/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/dstruct/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/lm/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/flm/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/lattice/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
cd ..; /home/kiit/srilm/sbin/make-standard-directories
make ../obj/i686-m64/STAMP ../bin/i686-m64/STAMP
make[3]: Entering directory `/home/kiit/srilm/utils/src'
make[3]: `../obj/i686-m64/STAMP' is up to date.
make[3]: `../bin/i686-m64/STAMP' is up to date.
make[3]: Leaving directory `/home/kiit/srilm/utils/src'
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make release-headers
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 release-headers) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
make[2]: Nothing to be done for `release-headers'.
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make depend
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 depend) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
rm -f Dependencies.i686-m64
gcc -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -Wimplicit-int    -I. -I../../include -fPIC -MM  ./option.c ./zio.c ./fcheck.c ./fake-rand48.c ./version.c ./ztest.c | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
g++ -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -DINSTANTIATE_TEMPLATES    -I. -I../../include -fPIC -MM  ./Debug.cc ./File.cc ./MStringTokUtil.cc ./testFile.cc | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 "" ztest testFile | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
rm -f Dependencies.i686-m64
gcc -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -Wimplicit-int    -I. -I../../include -fPIC -MM  ./qsort.c ./BlockMalloc.c ./maxalloc.c | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
g++ -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -DINSTANTIATE_TEMPLATES    -I. -I../../include -fPIC -MM  ./MemStats.cc ./LHashTrie.cc ./SArrayTrie.cc ./Array.cc ./IntervalHeap.cc ./Map.cc ./SArray.cc ./LHash.cc ./Map2.cc ./Trie.cc ./CachedMem.cc ./testArray.cc ./testMap.cc ./benchHash.cc ./testHash.cc ./testSizes.cc ./testCachedMem.cc ./testBlockMalloc.cc | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 "" maxalloc testArray testMap benchHash testHash testSizes testCachedMem testBlockMalloc | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
rm -f Dependencies.i686-m64
gcc -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -Wimplicit-int    -I. -I../../include -fPIC -MM  ./matherr.c | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
g++ -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -DINSTANTIATE_TEMPLATES    -I. -I../../include -fPIC -MM  ./Prob.cc ./Counts.cc ./XCount.cc ./Vocab.cc ./VocabMap.cc ./VocabMultiMap.cc ./VocabDistance.cc ./SubVocab.cc ./MultiwordVocab.cc ./TextStats.cc ./LM.cc ./LMClient.cc ./LMStats.cc ./RefList.cc ./Bleu.cc ./NBest.cc ./NBestSet.cc ./NgramLM.cc ./NgramStatsInt.cc ./NgramStatsShort.cc ./NgramStatsLong.cc ./NgramStatsLongLong.cc ./NgramStatsFloat.cc ./NgramStatsDouble.cc ./NgramStatsXCount.cc ./NgramCountLM.cc ./Discount.cc ./ClassNgram.cc ./SimpleClassNgram.cc ./DFNgram.cc ./SkipNgram.cc ./HiddenNgram.cc ./HiddenSNgram.cc ./VarNgram.cc ./DecipherNgram.cc ./TaggedVocab.cc ./TaggedNgram.cc ./TaggedNgramStats.cc ./StopNgram.cc ./StopNgramStats.cc ./MultiwordLM.cc ./NonzeroLM.cc ./BayesMix.cc ./LoglinearMix.cc ./AdaptiveMix.cc ./AdaptiveMarginals.cc ./CacheLM.cc ./DynamicLM.cc ./HMMofNgrams.cc ./WordAlign.cc ./WordLattice.cc ./WordMesh.cc ./simpleTrigram.cc ./NgramStats.cc ./Trellis.cc ./testBinaryCounts.cc ./testHash.cc ./testProb.cc ./testXCount.cc ./testParseFloat.cc ./testVocabDistance.cc ./testNgram.cc ./testNgramAlloc.cc ./testMultiReadLM.cc ./hoeffding.cc ./tolower.cc ./testLattice.cc ./testError.cc ./testNBest.cc ./testMix.cc ./ngram.cc ./ngram-count.cc ./ngram-merge.cc ./ngram-class.cc ./disambig.cc ./anti-ngram.cc ./nbest-lattice.cc ./nbest-mix.cc ./nbest-optimize.cc ./nbest-pron-score.cc ./segment.cc ./segment-nbest.cc ./hidden-ngram.cc ./multi-ngram.cc | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 "" testBinaryCounts testHash testProb testXCount testParseFloat testVocabDistance testNgram testNgramAlloc testMultiReadLM hoeffding tolower testLattice testError testNBest testMix ngram ngram-count ngram-merge ngram-class disambig anti-ngram nbest-lattice nbest-mix nbest-optimize nbest-pron-score segment segment-nbest hidden-ngram multi-ngram | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
rm -f Dependencies.i686-m64
g++ -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -DINSTANTIATE_TEMPLATES    -I. -I../../include -fPIC -MM  ./FDiscount.cc ./FNgramStats.cc ./FNgramStatsInt.cc ./FNgramSpecs.cc ./FNgramSpecsInt.cc ./FactoredVocab.cc ./FNgramLM.cc ./ProductVocab.cc ./ProductNgram.cc ./wmatrix.cc ./pngram.cc ./fngram-count.cc ./fngram.cc | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 "" pngram fngram-count fngram  | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
rm -f Dependencies.i686-m64
g++ -march=athlon64 -m64 -Wall -Wno-unused-variable -Wno-uninitialized -DINSTANTIATE_TEMPLATES    -I. -I../../include -fPIC -MM  ./Lattice.cc ./LatticeAlign.cc ./LatticeExpand.cc ./LatticeIndex.cc ./LatticeNBest.cc ./LatticeNgrams.cc ./LatticeReduce.cc ./HTKLattice.cc ./LatticeLM.cc ./LatticeDecode.cc ./testLattice.cc ./lattice-tool.cc | sed -e "s&^\([^ ]\)&../obj/i686-m64"'$(OBJ_OPTION)'"/\1&g" -e "s&\.o&.o&g" >> Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 "" testLattice  lattice-tool | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
rm -f Dependencies.i686-m64
/home/kiit/srilm/sbin/generate-program-dependencies ../bin/i686-m64 ../obj/i686-m64 ""  | sed -e "s&\.o&.o&g" >> Dependencies.i686-m64
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make release-libraries
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 release-libraries) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
make[2]: Nothing to be done for `release-libraries'.
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make release-programs
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 release-programs) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
make[2]: Nothing to be done for `release-programs'.
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/maxalloc ../../bin/i686-m64
WARNING: creating directory ../../bin/i686-m64
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/ngram ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/ngram-count ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/ngram-merge ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/ngram-class ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/disambig ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/anti-ngram ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-lattice ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-mix ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-optimize ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-pron-score ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/segment ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/segment-nbest ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/hidden-ngram ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/multi-ngram ../../bin/i686-m64
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/fngram-count ../../bin/i686-m64
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/fngram ../../bin/i686-m64
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/lattice-tool ../../bin/i686-m64
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' add-classes-to-pfsg.gawk >../bin/i686-m64/add-classes-to-pfsg.new
mv ../bin/i686-m64/add-classes-to-pfsg.new ../bin/i686-m64/add-classes-to-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/add-classes-to-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' add-dummy-bows.gawk >../bin/i686-m64/add-dummy-bows.new
mv ../bin/i686-m64/add-dummy-bows.new ../bin/i686-m64/add-dummy-bows
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/add-dummy-bows ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' add-pauses-to-pfsg.gawk >../bin/i686-m64/add-pauses-to-pfsg.new
mv ../bin/i686-m64/add-pauses-to-pfsg.new ../bin/i686-m64/add-pauses-to-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/add-pauses-to-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' add-ppls.gawk >../bin/i686-m64/add-ppls.new
mv ../bin/i686-m64/add-ppls.new ../bin/i686-m64/add-ppls
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/add-ppls ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' bytelog-to-log10.gawk >../bin/i686-m64/bytelog-to-log10.new
mv ../bin/i686-m64/bytelog-to-log10.new ../bin/i686-m64/bytelog-to-log10
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/bytelog-to-log10 ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' classes-to-fsm.gawk >../bin/i686-m64/classes-to-fsm.new
mv ../bin/i686-m64/classes-to-fsm.new ../bin/i686-m64/classes-to-fsm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/classes-to-fsm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' combine-acoustic-scores.gawk >../bin/i686-m64/combine-acoustic-scores.new
mv ../bin/i686-m64/combine-acoustic-scores.new ../bin/i686-m64/combine-acoustic-scores
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/combine-acoustic-scores ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' combine-rover-controls.gawk >../bin/i686-m64/combine-rover-controls.new
mv ../bin/i686-m64/combine-rover-controls.new ../bin/i686-m64/combine-rover-controls
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/combine-rover-controls ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' compare-ppls.gawk >../bin/i686-m64/compare-ppls.new
mv ../bin/i686-m64/compare-ppls.new ../bin/i686-m64/compare-ppls
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/compare-ppls ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' compute-best-mix.gawk >../bin/i686-m64/compute-best-mix.new
mv ../bin/i686-m64/compute-best-mix.new ../bin/i686-m64/compute-best-mix
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/compute-best-mix ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' compute-best-rover-mix.gawk >../bin/i686-m64/compute-best-rover-mix.new
mv ../bin/i686-m64/compute-best-rover-mix.new ../bin/i686-m64/compute-best-rover-mix
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/compute-best-rover-mix ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' compute-best-sentence-mix.gawk >../bin/i686-m64/compute-best-sentence-mix.new
mv ../bin/i686-m64/compute-best-sentence-mix.new ../bin/i686-m64/compute-best-sentence-mix
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/compute-best-sentence-mix ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' compute-oov-rate.gawk >../bin/i686-m64/compute-oov-rate.new
mv ../bin/i686-m64/compute-oov-rate.new ../bin/i686-m64/compute-oov-rate
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/compute-oov-rate ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' context-ngrams.gawk >../bin/i686-m64/context-ngrams.new
mv ../bin/i686-m64/context-ngrams.new ../bin/i686-m64/context-ngrams
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/context-ngrams ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' continuous-ngram-count.gawk >../bin/i686-m64/continuous-ngram-count.new
mv ../bin/i686-m64/continuous-ngram-count.new ../bin/i686-m64/continuous-ngram-count
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/continuous-ngram-count ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' extract-skip-probs.gawk >../bin/i686-m64/extract-skip-probs.new
mv ../bin/i686-m64/extract-skip-probs.new ../bin/i686-m64/extract-skip-probs
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/extract-skip-probs ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' filter-event-counts.gawk >../bin/i686-m64/filter-event-counts.new
mv ../bin/i686-m64/filter-event-counts.new ../bin/i686-m64/filter-event-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/filter-event-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' find-reference-posteriors.gawk >../bin/i686-m64/find-reference-posteriors.new
mv ../bin/i686-m64/find-reference-posteriors.new ../bin/i686-m64/find-reference-posteriors
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/find-reference-posteriors ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' fix-ctm.gawk >../bin/i686-m64/fix-ctm.new
mv ../bin/i686-m64/fix-ctm.new ../bin/i686-m64/fix-ctm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/fix-ctm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' fsm-to-pfsg.gawk >../bin/i686-m64/fsm-to-pfsg.new
mv ../bin/i686-m64/fsm-to-pfsg.new ../bin/i686-m64/fsm-to-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/fsm-to-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' get-gt-counts.gawk >../bin/i686-m64/get-gt-counts.new
mv ../bin/i686-m64/get-gt-counts.new ../bin/i686-m64/get-gt-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/get-gt-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' get-unigram-probs.gawk >../bin/i686-m64/get-unigram-probs.new
mv ../bin/i686-m64/get-unigram-probs.new ../bin/i686-m64/get-unigram-probs
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/get-unigram-probs ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' hits-from-log.gawk >../bin/i686-m64/hits-from-log.new
mv ../bin/i686-m64/hits-from-log.new ../bin/i686-m64/hits-from-log
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/hits-from-log ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' log10-to-bytelog.gawk >../bin/i686-m64/log10-to-bytelog.new
mv ../bin/i686-m64/log10-to-bytelog.new ../bin/i686-m64/log10-to-bytelog
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/log10-to-bytelog ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-abs-discount.gawk >../bin/i686-m64/make-abs-discount.new
mv ../bin/i686-m64/make-abs-discount.new ../bin/i686-m64/make-abs-discount
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-abs-discount ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-diacritic-map.gawk >../bin/i686-m64/make-diacritic-map.new
mv ../bin/i686-m64/make-diacritic-map.new ../bin/i686-m64/make-diacritic-map
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-diacritic-map ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-google-ngrams.gawk >../bin/i686-m64/make-google-ngrams.new
mv ../bin/i686-m64/make-google-ngrams.new ../bin/i686-m64/make-google-ngrams
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-google-ngrams ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-gt-discounts.gawk >../bin/i686-m64/make-gt-discounts.new
mv ../bin/i686-m64/make-gt-discounts.new ../bin/i686-m64/make-gt-discounts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-gt-discounts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-kn-discounts.gawk >../bin/i686-m64/make-kn-discounts.new
mv ../bin/i686-m64/make-kn-discounts.new ../bin/i686-m64/make-kn-discounts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-kn-discounts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-kn-counts.gawk >../bin/i686-m64/make-kn-counts.new
mv ../bin/i686-m64/make-kn-counts.new ../bin/i686-m64/make-kn-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-kn-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-hiddens-lm.gawk >../bin/i686-m64/make-hiddens-lm.new
mv ../bin/i686-m64/make-hiddens-lm.new ../bin/i686-m64/make-hiddens-lm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-hiddens-lm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-lm-subset.gawk >../bin/i686-m64/make-lm-subset.new
mv ../bin/i686-m64/make-lm-subset.new ../bin/i686-m64/make-lm-subset
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-lm-subset ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-nbest-pfsg.gawk >../bin/i686-m64/make-nbest-pfsg.new
mv ../bin/i686-m64/make-nbest-pfsg.new ../bin/i686-m64/make-nbest-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-nbest-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-ngram-pfsg.gawk >../bin/i686-m64/make-ngram-pfsg.new
mv ../bin/i686-m64/make-ngram-pfsg.new ../bin/i686-m64/make-ngram-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-ngram-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' make-sub-lm.gawk >../bin/i686-m64/make-sub-lm.new
mv ../bin/i686-m64/make-sub-lm.new ../bin/i686-m64/make-sub-lm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/make-sub-lm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' metadb.gawk >../bin/i686-m64/metadb.new
mv ../bin/i686-m64/metadb.new ../bin/i686-m64/metadb
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/metadb ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' sort-lm.gawk >../bin/i686-m64/sort-lm.new
mv ../bin/i686-m64/sort-lm.new ../bin/i686-m64/sort-lm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/sort-lm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' reverse-lm.gawk >../bin/i686-m64/reverse-lm.new
mv ../bin/i686-m64/reverse-lm.new ../bin/i686-m64/reverse-lm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/reverse-lm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' merge-nbest.gawk >../bin/i686-m64/merge-nbest.new
mv ../bin/i686-m64/merge-nbest.new ../bin/i686-m64/merge-nbest
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/merge-nbest ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' nbest-posteriors.gawk >../bin/i686-m64/nbest-posteriors.new
mv ../bin/i686-m64/nbest-posteriors.new ../bin/i686-m64/nbest-posteriors
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-posteriors ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' nbest2-to-nbest1.gawk >../bin/i686-m64/nbest2-to-nbest1.new
mv ../bin/i686-m64/nbest2-to-nbest1.new ../bin/i686-m64/nbest2-to-nbest1
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest2-to-nbest1 ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' nbest-vocab.gawk >../bin/i686-m64/nbest-vocab.new
mv ../bin/i686-m64/nbest-vocab.new ../bin/i686-m64/nbest-vocab
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/nbest-vocab ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' pfsg-to-dot.gawk >../bin/i686-m64/pfsg-to-dot.new
mv ../bin/i686-m64/pfsg-to-dot.new ../bin/i686-m64/pfsg-to-dot
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/pfsg-to-dot ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' pfsg-to-fsm.gawk >../bin/i686-m64/pfsg-to-fsm.new
mv ../bin/i686-m64/pfsg-to-fsm.new ../bin/i686-m64/pfsg-to-fsm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/pfsg-to-fsm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' pfsg-vocab.gawk >../bin/i686-m64/pfsg-vocab.new
mv ../bin/i686-m64/pfsg-vocab.new ../bin/i686-m64/pfsg-vocab
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/pfsg-vocab ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' htklat-vocab.gawk >../bin/i686-m64/htklat-vocab.new
mv ../bin/i686-m64/htklat-vocab.new ../bin/i686-m64/htklat-vocab
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/htklat-vocab ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' ppl-from-log.gawk >../bin/i686-m64/ppl-from-log.new
mv ../bin/i686-m64/ppl-from-log.new ../bin/i686-m64/ppl-from-log
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/ppl-from-log ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' remove-lowprob-ngrams.gawk >../bin/i686-m64/remove-lowprob-ngrams.new
mv ../bin/i686-m64/remove-lowprob-ngrams.new ../bin/i686-m64/remove-lowprob-ngrams
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/remove-lowprob-ngrams ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' replace-words-with-classes.gawk >../bin/i686-m64/replace-words-with-classes.new
mv ../bin/i686-m64/replace-words-with-classes.new ../bin/i686-m64/replace-words-with-classes
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/replace-words-with-classes ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' reverse-text.gawk >../bin/i686-m64/reverse-text.new
mv ../bin/i686-m64/reverse-text.new ../bin/i686-m64/reverse-text
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/reverse-text ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' reverse-ngram-counts.gawk >../bin/i686-m64/reverse-ngram-counts.new
mv ../bin/i686-m64/reverse-ngram-counts.new ../bin/i686-m64/reverse-ngram-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/reverse-ngram-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' sentid-to-sclite.gawk >../bin/i686-m64/sentid-to-sclite.new
mv ../bin/i686-m64/sentid-to-sclite.new ../bin/i686-m64/sentid-to-sclite
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/sentid-to-sclite ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' sentid-to-ctm.gawk >../bin/i686-m64/sentid-to-ctm.new
mv ../bin/i686-m64/sentid-to-ctm.new ../bin/i686-m64/sentid-to-ctm
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/sentid-to-ctm ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' split-tagged-ngrams.gawk >../bin/i686-m64/split-tagged-ngrams.new
mv ../bin/i686-m64/split-tagged-ngrams.new ../bin/i686-m64/split-tagged-ngrams
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/split-tagged-ngrams ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' subset-context-ngrams.gawk >../bin/i686-m64/subset-context-ngrams.new
mv ../bin/i686-m64/subset-context-ngrams.new ../bin/i686-m64/subset-context-ngrams
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/subset-context-ngrams ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' subtract-ppls.gawk >../bin/i686-m64/subtract-ppls.new
mv ../bin/i686-m64/subtract-ppls.new ../bin/i686-m64/subtract-ppls
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/subtract-ppls ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' tolower-ngram-counts.gawk >../bin/i686-m64/tolower-ngram-counts.new
mv ../bin/i686-m64/tolower-ngram-counts.new ../bin/i686-m64/tolower-ngram-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/tolower-ngram-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' uniform-classes.gawk >../bin/i686-m64/uniform-classes.new
mv ../bin/i686-m64/uniform-classes.new ../bin/i686-m64/uniform-classes
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/uniform-classes ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' uniq-ngram-counts.gawk >../bin/i686-m64/uniq-ngram-counts.new
mv ../bin/i686-m64/uniq-ngram-counts.new ../bin/i686-m64/uniq-ngram-counts
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/uniq-ngram-counts ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' vp2text.gawk >../bin/i686-m64/vp2text.new
mv ../bin/i686-m64/vp2text.new ../bin/i686-m64/vp2text
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/vp2text ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' wlat-to-dot.gawk >../bin/i686-m64/wlat-to-dot.new
mv ../bin/i686-m64/wlat-to-dot.new ../bin/i686-m64/wlat-to-dot
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/wlat-to-dot ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' wlat-to-pfsg.gawk >../bin/i686-m64/wlat-to-pfsg.new
mv ../bin/i686-m64/wlat-to-pfsg.new ../bin/i686-m64/wlat-to-pfsg
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/wlat-to-pfsg ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' wlat-stats.gawk >../bin/i686-m64/wlat-stats.new
mv ../bin/i686-m64/wlat-stats.new ../bin/i686-m64/wlat-stats
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/wlat-stats ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' wordlat-to-lisp.gawk >../bin/i686-m64/wordlat-to-lisp.new
mv ../bin/i686-m64/wordlat-to-lisp.new ../bin/i686-m64/wordlat-to-lisp
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/wordlat-to-lisp ../../bin/i686-m64
sed -e '1s,/usr/local/bin/gawk,/usr/bin/awk,' prettify.gawk >../bin/i686-m64/prettify.new
mv ../bin/i686-m64/prettify.new ../bin/i686-m64/prettify
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/prettify ../../bin/i686-m64
sed -e '1s,/usr/local/bin/perl,/usr/local/bin/perl,' select-vocab.pl >../bin/i686-m64/select-vocab.new
mv ../bin/i686-m64/select-vocab.new ../bin/i686-m64/select-vocab
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/select-vocab ../../bin/i686-m64
sed -e '1s,/usr/local/bin/perl,/usr/local/bin/perl,' cumbin.pl >../bin/i686-m64/cumbin.new
mv ../bin/i686-m64/cumbin.new ../bin/i686-m64/cumbin
/home/kiit/srilm/sbin/decipher-install 0555 ../bin/i686-m64/cumbin ../../bin/i686-m64
rm ../bin/i686-m64/merge-nbest ../bin/i686-m64/reverse-text ../bin/i686-m64/hits-from-log ../bin/i686-m64/pfsg-vocab ../bin/i686-m64/sentid-to-ctm ../bin/i686-m64/reverse-lm ../bin/i686-m64/add-ppls ../bin/i686-m64/sort-lm ../bin/i686-m64/sentid-to-sclite ../bin/i686-m64/extract-skip-probs ../bin/i686-m64/fsm-to-pfsg ../bin/i686-m64/make-diacritic-map ../bin/i686-m64/wlat-stats ../bin/i686-m64/compute-best-rover-mix ../bin/i686-m64/tolower-ngram-counts ../bin/i686-m64/bytelog-to-log10 ../bin/i686-m64/make-hiddens-lm ../bin/i686-m64/cumbin ../bin/i686-m64/add-classes-to-pfsg ../bin/i686-m64/wordlat-to-lisp ../bin/i686-m64/pfsg-to-dot ../bin/i686-m64/make-sub-lm ../bin/i686-m64/prettify ../bin/i686-m64/wlat-to-dot ../bin/i686-m64/combine-rover-controls ../bin/i686-m64/filter-event-counts ../bin/i686-m64/wlat-to-pfsg ../bin/i686-m64/remove-lowprob-ngrams ../bin/i686-m64/make-nbest-pfsg ../bin/i686-m64/make-gt-discounts ../bin/i686-m64/nbest2-to-nbest1 ../bin/i686-m64/select-vocab ../bin/i686-m64/classes-to-fsm ../bin/i686-m64/fix-ctm ../bin/i686-m64/subtract-ppls ../bin/i686-m64/compute-best-mix ../bin/i686-m64/get-gt-counts ../bin/i686-m64/make-lm-subset ../bin/i686-m64/log10-to-bytelog ../bin/i686-m64/split-tagged-ngrams ../bin/i686-m64/uniq-ngram-counts ../bin/i686-m64/context-ngrams ../bin/i686-m64/make-abs-discount ../bin/i686-m64/replace-words-with-classes ../bin/i686-m64/pfsg-to-fsm ../bin/i686-m64/find-reference-posteriors ../bin/i686-m64/combine-acoustic-scores ../bin/i686-m64/compute-best-sentence-mix ../bin/i686-m64/make-kn-discounts ../bin/i686-m64/add-dummy-bows ../bin/i686-m64/vp2text ../bin/i686-m64/continuous-ngram-count ../bin/i686-m64/make-ngram-pfsg ../bin/i686-m64/ppl-from-log ../bin/i686-m64/htklat-vocab ../bin/i686-m64/uniform-classes ../bin/i686-m64/nbest-vocab ../bin/i686-m64/compute-oov-rate ../bin/i686-m64/make-kn-counts ../bin/i686-m64/reverse-ngram-counts ../bin/i686-m64/compare-ppls ../bin/i686-m64/add-pauses-to-pfsg ../bin/i686-m64/metadb ../bin/i686-m64/make-google-ngrams ../bin/i686-m64/subset-context-ngrams ../bin/i686-m64/get-unigram-probs ../bin/i686-m64/nbest-posteriors
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make release-scripts
make[1]: Entering directory `/home/kiit/srilm'
for subdir in misc dstruct lm flm lattice utils; do \
		(cd $subdir/src; make SRILM=/home/kiit/srilm MACHINE_TYPE=i686-m64 OPTION= MAKE_PIC=1 release-scripts) || exit 1; \
	done
make[2]: Entering directory `/home/kiit/srilm/misc/src'
make[2]: Nothing to be done for `release-scripts'.
make[2]: Leaving directory `/home/kiit/srilm/misc/src'
make[2]: Entering directory `/home/kiit/srilm/dstruct/src'
make[2]: Nothing to be done for `release-scripts'.
make[2]: Leaving directory `/home/kiit/srilm/dstruct/src'
make[2]: Entering directory `/home/kiit/srilm/lm/src'
make[2]: Nothing to be done for `release-scripts'.
make[2]: Leaving directory `/home/kiit/srilm/lm/src'
make[2]: Entering directory `/home/kiit/srilm/flm/src'
make[2]: Nothing to be done for `release-scripts'.
make[2]: Leaving directory `/home/kiit/srilm/flm/src'
make[2]: Entering directory `/home/kiit/srilm/lattice/src'
make[2]: Nothing to be done for `release-scripts'.
make[2]: Leaving directory `/home/kiit/srilm/lattice/src'
make[2]: Entering directory `/home/kiit/srilm/utils/src'
/home/kiit/srilm/sbin/decipher-install 0555 change-lm-vocab ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 empty-sentence-lm ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 rescore-decipher ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 rescore-acoustic ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 rescore-reweight ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 rescore-minimize-wer ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 make-batch-counts ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 merge-batch-counts ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 make-big-lm ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 make-multiword-pfsg ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 pfsg-from-ngram ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 nbest-error ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 nbest-rover ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 align-with-tags ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 compute-sclite ../../bin
/home/kiit/srilm/sbin/decipher-install 0555 compare-sclite ../../bin
make[2]: Leaving directory `/home/kiit/srilm/utils/src'
make[1]: Leaving directory `/home/kiit/srilm'
make: Leaving directory `/home/kiit/srilm'

From stolcke at icsi.berkeley.edu  Thu Mar 21 00:07:04 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Thu, 21 Mar 2013 00:07:04 -0700
Subject: [SRILM User List] Fwd: Problem- "cannot execute binary file"
In-Reply-To: <CAH7UGAo4mJ7nJtQi-GgiOSjdmdhKe68dALNC0h7eQkNhQ6mrjA@mail.gmail.com>
References: <CAH7UGApf6ssdVez+wxXwdPLhbgHMj4ifc4T5RRJ3gY74skLBhA@mail.gmail.com>
	<CAH7UGAo4mJ7nJtQi-GgiOSjdmdhKe68dALNC0h7eQkNhQ6mrjA@mail.gmail.com>
Message-ID: <514AB198.2090708@icsi.berkeley.edu>

On 3/20/2013 11:53 PM, SHUBHAMAY SEN wrote:
>
>
> ---------- Forwarded message ----------
> From: *SHUBHAMAY SEN* <shubhamay.sen at gmail.com 
> <mailto:shubhamay.sen at gmail.com>>
> Date: Thu, Mar 21, 2013 at 12:14 PM
> Subject: Problem- "cannot execute binary file"
> To: srilm-user at speech.sri.com <mailto:srilm-user at speech.sri.com>
>
>
> Hi,
>       I have a 64 bit machine.
>
> kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$ uname -a
> Linux kiit-ThinkCentre-M72e 3.5.0-21-generic #32-Ubuntu SMP Tue Dec 11 
> 18:52:46 UTC 2012 i686 i686 i686 GNU/Linux
Note the "i686", that means your machine or OS are 32bit.

>
> I have compiled srilm using "make World MAKE_PIC=1 
> MACHINE_TYPE=i686-m64 -C $SRILM" command, output is attached.

but you're building 64bit binaries...
>
> but when I am executing the below command, it is showing the following 
> message.
>
> kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$ 
> $SRILM/bin/*/ngram-count -order 3
> -interpolate -kndiscount -unk -text 
> data/training/news-commentary.tok.en -lm lm/news.lm.gz
>
> bash: /home/kiit/srilm/bin/i686-m64/ngram-count: cannot execute binary 
> file

... therefore they will not run.

Build with MACHINE_TYPE=i686 .

Andreas


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130321/6a2bf881/attachment.html>

From stolcke at icsi.berkeley.edu  Thu Mar 21 10:57:19 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Thu, 21 Mar 2013 10:57:19 -0700
Subject: [SRILM User List] Fwd: Problem- "cannot execute binary file"
In-Reply-To: <CAH7UGApHAj9VkCM01=aiAgmgv38SXi1BotAWAZsrOhvGEAOLkQ@mail.gmail.com>
References: <CAH7UGApf6ssdVez+wxXwdPLhbgHMj4ifc4T5RRJ3gY74skLBhA@mail.gmail.com>
	<CAH7UGAo4mJ7nJtQi-GgiOSjdmdhKe68dALNC0h7eQkNhQ6mrjA@mail.gmail.com>
	<514AB198.2090708@icsi.berkeley.edu>
	<CAH7UGAo_N5=6Z88RU_YWOsDbrL1==SW1qkw5zgERrZ2rkqF=_Q@mail.gmail.com>
	<514B2EB7.5030007@icsi.berkeley.edu>
	<CAH7UGApHAj9VkCM01=aiAgmgv38SXi1BotAWAZsrOhvGEAOLkQ@mail.gmail.com>
Message-ID: <514B49FF.3000308@icsi.berkeley.edu>

Glad to hear it!

Andreas

On 3/21/2013 9:18 AM, SHUBHAMAY SEN wrote:
> Actually the system is running properly now
> we have done the modifications in srilm/common/Makefile.machin.i686
> as below
>  TCL_INCLUDE =
>  TCL_LIBRARY =
>  NO_TCL = X
>
> Thanks
>
>
> On Thu, Mar 21, 2013 at 9:30 PM, Andreas Stolcke 
> <stolcke at icsi.berkeley.edu <mailto:stolcke at icsi.berkeley.edu>> wrote:
>
>     Capture the output of make and send it.
>
>     Andreas
>
>
>     On 3/21/2013 12:41 AM, SHUBHAMAY SEN wrote:
>>
>>
>>     On Thu, Mar 21, 2013 at 12:37 PM, Andreas Stolcke
>>     <stolcke at icsi.berkeley.edu <mailto:stolcke at icsi.berkeley.edu>> wrote:
>>
>>         On 3/20/2013 11:53 PM, SHUBHAMAY SEN wrote:
>>>
>>>
>>>         ---------- Forwarded message ----------
>>>         From: *SHUBHAMAY SEN* <shubhamay.sen at gmail.com
>>>         <mailto:shubhamay.sen at gmail.com>>
>>>         Date: Thu, Mar 21, 2013 at 12:14 PM
>>>         Subject: Problem- "cannot execute binary file"
>>>         To: srilm-user at speech.sri.com <mailto:srilm-user at speech.sri.com>
>>>
>>>
>>>         Hi,
>>>               I have a 64 bit machine.
>>>
>>>         kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$
>>>         <mailto:kiit at kiit-ThinkCentre-M72e:%7E/phrasal.Beta4/work$>
>>>         uname -a
>>>         Linux kiit-ThinkCentre-M72e 3.5.0-21-generic #32-Ubuntu SMP
>>>         Tue Dec 11 18:52:46 UTC 2012 i686 i686 i686 GNU/Linux
>>         Note the "i686", that means your machine or OS are 32bit.
>>
>>     Thanks.
>>         I have successfully compiled with the following command.
>>                   make World MAKE_PIC=1 MACHINE_TYPE=i686 -C $SRILM
>>     NO_TCL=X
>>        But the ngram-count file is not generated in bin/i686
>>     directory. The other files are successfully generated.
>>     Please suggest.
>>
>>
>>>
>>>         I have compiled srilm using "make World MAKE_PIC=1
>>>         MACHINE_TYPE=i686-m64 -C $SRILM" command, output is attached.
>>
>>         but you're building 64bit binaries...
>>
>>>
>>>         but when I am executing the below command, it is showing the
>>>         following message.
>>>
>>>         kiit at kiit-ThinkCentre-M72e:~/phrasal.Beta4/work$
>>>         <mailto:kiit at kiit-ThinkCentre-M72e:%7E/phrasal.Beta4/work$>
>>>         $SRILM/bin/*/ngram-count -order 3
>>>         -interpolate -kndiscount -unk -text
>>>         data/training/news-commentary.tok.en -lm lm/news.lm.gz
>>>
>>>         bash: /home/kiit/srilm/bin/i686-m64/ngram-count: cannot
>>>         execute binary file
>>
>>         ... therefore they will not run.
>>
>>         Build with MACHINE_TYPE=i686 .
>>
>>         Andreas
>>
>>
>>
>>
>>
>>     -- 
>>     ---------------------------------------------------------------------
>>     Regards:
>>     Shubhamay Sen
>>     M. Tech CSE
>>     KIIT University, BBSR,
>>     Odisha, India
>>     mob- 09778772782 
>
>
>
>
> -- 
> ---------------------------------------------------------------------
> Regards:
> Shubhamay Sen
> M. Tech CSE
> KIIT University, BBSR,
> Odisha, India
> mob- 09778772782

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.speech.sri.com/pipermail/srilm-user/attachments/20130321/573bc816/attachment.html>

From S.N.Maijers at student.ru.nl  Wed Mar 27 05:44:08 2013
From: S.N.Maijers at student.ru.nl (Sander Maijers)
Date: Wed, 27 Mar 2013 13:44:08 +0100
Subject: [SRILM User List] Detailed description of ngram-count's -skip option
Message-ID: <5152E998.3090607@student.ru.nl>

Hello everyone,

Where can I find more detailed information about the word skipping 
algorithm provided by ngram-count? Thus far I found this:

"Skip language models ? In this LM, words in the history are 
probabilistically skipped, allowing more distant words to take their 
places. The skipping probabilities associated with each word are 
estimated using expectation maximization." (In SRILM ? AN EXTENSIBLE 
LANGUAGE MODELING TOOLKIT, Stolcke, 2002)

As I want to refer to this I prefer published scientific work authored 
by Andreas Stolcke.

Best,
Sander Maijers

From stolcke at icsi.berkeley.edu  Thu Mar 28 10:03:11 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Thu, 28 Mar 2013 10:03:11 -0700
Subject: [SRILM User List] Detailed description of ngram-count's -skip
 option
In-Reply-To: <5152E998.3090607@student.ru.nl>
References: <5152E998.3090607@student.ru.nl>
Message-ID: <515477CF.80602@icsi.berkeley.edu>

On 3/27/2013 5:44 AM, Sander Maijers wrote:
> Hello everyone,
>
> Where can I find more detailed information about the word skipping 
> algorithm provided by ngram-count? Thus far I found this:
>
> "Skip language models ? In this LM, words in the history are 
> probabilistically skipped, allowing more distant words to take their 
> places. The skipping probabilities associated with each word are 
> estimated using expectation maximization." (In SRILM ? AN EXTENSIBLE 
> LANGUAGE MODELING TOOLKIT, Stolcke, 2002)
>
> As I want to refer to this I prefer published scientific work authored 
> by Andreas Stolcke.

The last statement makes me uncomfortable.  Skip ngrams are a variant of 
"distant ngram" models that you can find in the literature prior to me 
writing this particular implementation.  I am having trouble finding a 
good reference prior to 1995, but Roni Rosenfeld's 1994 thesis certainly 
had similar ideas, though in the context of maxent modeling.

The SkipNgram model is essentially an interpolation between a straight 
ngram P(w_n | w_{n-1} w_{n-2} ...)  and another ngram model where the 
preceding word is skipped: P(w_n | w_{n-2} P_{n-1} where the 
interpolation weight is a function of the skipped word w_{n-1}.
So you have a "skipping probability" associated with each word, and that 
is estimated in a straightforward way using EM.  You can read the code 
for the details, it should be pretty easy to follow.

Andreas


From stolcke at icsi.berkeley.edu  Fri Mar 29 09:06:18 2013
From: stolcke at icsi.berkeley.edu (Andreas Stolcke)
Date: Fri, 29 Mar 2013 09:06:18 -0700
Subject: [SRILM User List] LM of phonemes strings
In-Reply-To: <51549CA5.2080506@cenatav.co.cu>
References: <51549CA5.2080506@cenatav.co.cu>
Message-ID: <5155BBFA.7060700@icsi.berkeley.edu>

On 3/28/2013 12:40 PM, Ana Montalvo Bereau wrote:
> Hello all, my name is Ana, I'm a beginner with srilm.
> My objective is to construct a language model to make spoken language 
> recognition.
> I'll use the output of a phoneme recognizer to train the LM, so my 
> doubt is if SRILM allows me to build models that estimate the prior 
> probabilities of phonemes strings rather than words strings.
> In case of positive answer, wich should be the procedure?
> thx in advance
> ana
Ana,

there is nothing really different about building phone-based language 
models.   In a phone recognizer the phone labels are treated just the 
same as the words in a word recognizer, and the same is true of the 
LM.     You just prepare a corpus of phone labels separated by white 
space (don't forget the phone representing "pause" or nonspeech), then 
use the ngram-count in the usual way to train an LM.

A minor difference is that because the vocabulary is small and finite, 
different smoothing methods might work best.  For example, Witten-Bell 
smoothing is a good choice for phone LMs in my experience.

Andreas