Software

From SpeechWiki

(Difference between revisions)
Jump to: navigation, search
Line 1: Line 1:
 +
=Statistical Speech Technology Group Software=
 +
 +
Our policy: everything we write is free on the web.  This wiki page is intended to be definitive, because anybody in the group can edit it to add their own software.  [http://www.isle.uiuc.edu/software/index.html] provides a spider-indexable backup copy.
 +
 +
All of our software is available via <a href="http://subversion.tigris.org/">subversion</a>, using login name "anon" with no password (hit "enter" when a password is requested).
 +
 +
* If you are using Windows,
 +
download [http://tortoisesvn.tigris.org/ TortoiseSVN]
 +
* If you are using linux, use the svn command interface, e.g.,
 +
 +
svn co svn://mickey.ifp.uiuc.edu/speechfileformats
 +
 +
<table border=2><tr>
 +
<tr><td>Learning</td></tr>
 +
<tr><td>Pronounce</td><td>Letters to phones using an HMM<br>
 +
<a href="http://mickey.ifp.uiuc.edu/speechWiki/index.php/Phonetic_Transcription_Tool">Description,</a>
 +
<a href="svn://mickey.ifp.uiuc.edu/pronounce">SVN archive,</a>
 +
<a href="http://mickey.ifp.uiuc.edu/speech/webpronounce/webpronounce.cgi">Demo
 +
</a>  (Arthur Kantor, 2007)</td></tr>
 +
 +
<tr><td>HDK</td><td>HTK-based Explicit-duration HMM<br>
 +
<a href="http://www.isle.uiuc.edu/pubs/2003/chen03interspeech.pdf">
 +
Description</a>;
 +
<a href="svn://mickey.ifp.uiuc.edu/HDK4_release">SVN repository</a>;
 +
<a href="http://www.isle.uiuc.edu/software/HDK4.tar.gz">TGZ archive</a>
 +
(Ken Chen, 2003)
 +
</td></tr>
 +
 +
<tr><td>Signal Processing</td></tr>
 +
<tr><td>PVTK</td><td>Extract HTK features as training vecs
 +
for libSVM, apply trained SVMs directly to feature files<br>
 +
<a href="svn://mickey.ifp.uiuc.edu/PVTK">SVN repository</a>;
 +
<a href="http://www.isle.uiuc.edu/software/PVTK2005May23.tgz">TGZ archive</a>
 +
(Sarah Borys 2008, Mark Hasegawa-Johnson 2005)
 +
</td></tr>
 +
<tr><td>VAD</td><td>Voice activity detector w/improved noise model<br>
 +
<a href="http://www.isle.uiuc.edu/pubs/2007/lee07dspincars.pdf">Description,</a>
 +
<a href="svn://mickey.ifp.uiuc.edu/lee_vad">SVN repository,</a>
 +
<a href="http://www.isle.uiuc.edu/software/lee_vad.m">lee_vad.m.</a>
 +
(Bowon Lee, 2007)</td></tr>
 +
 +
<tr><td>Computation</td></tr>
 +
<tr><td>GMTK Parallel</td>
 +
<td>Split GMTK commands into batch jobs for a cluster<br>
 +
<a href="http://mickey.ifp.uiuc.edu/speechWiki/index.php/GMTK_parallel_tools">
 +
Description,</a>
 +
<a href="svn://mickey.ifp.uiuc.edu/gmtkScripts/">SVN repository</a>.
 +
(Arthur Kantor, 2008)</td></tr>
 +
 +
<tr><td>HTK Parallel
 +
</td><td>
 +
Split an HTK command into batch jobs for a cluster (Bowon Lee, 2006)<br>
 +
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/cluster/HTK_parallel.htm">
 +
Description</a>;
 +
<a href="svn://mickey.ifp.uiuc.edu/HTK_parallel/">SVN repository</a>;
 +
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HCopy.pl">
 +
HCopy.pl</a>,
 +
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HVite.pl">
 +
HVite.pl</a>,
 +
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HERest.pl">
 +
HERest.pl</a>,
 +
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HResults.pl">
 +
HResults.pl</a>.</td></tr>
 +
 +
<tr><td>Data</td></tr>
 +
<tr><td>dtmfseg</td><td>Segment audio files at DTMF tones<br>
 +
<a href="svn://mickey.ifp.uiuc.edu/dtmfseg/">SVN repository</a>
 +
(Bowon Lee, 2006)</td></tr>
 +
 +
<tr><td>transcription tools</td><td>Convert transcription formats<br>
 +
<a href="svn://mickey.ifp.uiuc.edu/transcription_tools/">SVN repository</a>
 +
<a href="http://www.isle.uiuc.edu/software/transcription_tools2005May.tgz">
 +
TGZ archive</a>
 +
(Mark Hasegawa-Johnson, 2005)</td></tr>
 +
 +
<tr><td>speechfileformats</td><td>Read and write HTK files in matlab<br>
 +
<a href="svn://mickey.ifp.uiuc.edu/speechfileformats/">SVN repository</a>
 +
<a href="http://www.isle.uiuc.edu/software/speechfileformats.tgz">
 +
TGZ archive</a>
 +
(Mark Hasegawa-Johnson, 2004)</td></tr>
 +
 +
<tr><td>CTMRedit</td><td>Manually and automatically
 +
segment CT and MR image stacks<br>
 +
<a href="http://www.isle.uiuc.edu/pubs/1990s/hasegawa-johnson99embs.pdf">
 +
Description</a>;
 +
<a href="svn://mickey.ifp.uiuc.edu/CTMRedit/">SVN repository</a>
 +
(Mark Hasegawa-Johnson and Jul Cha, 1999)
 +
</td></tr>
 +
 +
</table>
 +
 +
=Phonetic Transcription Tool=
=Phonetic Transcription Tool=
Line 25: Line 117:
Written by Bowon Lee, 02/24/2006
Written by Bowon Lee, 02/24/2006
-
 
-
= Parallel GMTK =
 
-
[[GMTK parallel tools]]
 
-
 
-
= More software written at ISLE =
 
-
 
-
* Acoustic Model Training
 
-
** [http://www.isle.uiuc.edu/software/HDK4.tar.gz HDK]: HTK-based tools for training and testing explicit-duration hidden Markov models.  Contains makefiles and Windows binaries for the tools HDInit, HDrest, HDERest, HDHed,and HDVite.  Ken Chen, Last updated 5/2004.
 
-
 
-
* File Format Tools
 
-
** [http://www.isle.uiuc.edu/software/PVTK2005May23.tgz PVTK] Apply linear transforms to HTK files, or extract vectors in libSVM format.  Mark Hasegawa-Johnson, Last updated 5/2005.
 
-
** [http://www.isle.uiuc.edu/software/transcription_tools2005May.tgz Transcription tools]: Perl scripts for manipulating speech database transcriptions and word lattices. Mark Hasegawa-Johnson, last updated 5/2005.
 
-
** [http://www.isle.uiuc.edu/software/speechfileformats.tgz speechfileformats] Matlab tools for reading and writing HTK, MLF, and NIST files.
 
-
 
-
=See Also=
 
-
[[Data On Line]]
 

Revision as of 16:07, 4 November 2008

Statistical Speech Technology Group Software

Our policy: everything we write is free on the web. This wiki page is intended to be definitive, because anybody in the group can edit it to add their own software. [1] provides a spider-indexable backup copy.

All of our software is available via <a href="http://subversion.tigris.org/">subversion</a>, using login name "anon" with no password (hit "enter" when a password is requested).

  • If you are using Windows,

download TortoiseSVN

  • If you are using linux, use the svn command interface, e.g.,

svn co svn://mickey.ifp.uiuc.edu/speechfileformats

Learning
PronounceLetters to phones using an HMM

<a href="http://mickey.ifp.uiuc.edu/speechWiki/index.php/Phonetic_Transcription_Tool">Description,</a> <a href="svn://mickey.ifp.uiuc.edu/pronounce">SVN archive,</a> <a href="http://mickey.ifp.uiuc.edu/speech/webpronounce/webpronounce.cgi">Demo

</a> (Arthur Kantor, 2007)
HDKHTK-based Explicit-duration HMM

<a href="http://www.isle.uiuc.edu/pubs/2003/chen03interspeech.pdf"> Description</a>; <a href="svn://mickey.ifp.uiuc.edu/HDK4_release">SVN repository</a>; <a href="http://www.isle.uiuc.edu/software/HDK4.tar.gz">TGZ archive</a> (Ken Chen, 2003)

Signal Processing
PVTKExtract HTK features as training vecs

for libSVM, apply trained SVMs directly to feature files
<a href="svn://mickey.ifp.uiuc.edu/PVTK">SVN repository</a>; <a href="http://www.isle.uiuc.edu/software/PVTK2005May23.tgz">TGZ archive</a> (Sarah Borys 2008, Mark Hasegawa-Johnson 2005)

VADVoice activity detector w/improved noise model

<a href="http://www.isle.uiuc.edu/pubs/2007/lee07dspincars.pdf">Description,</a> <a href="svn://mickey.ifp.uiuc.edu/lee_vad">SVN repository,</a> <a href="http://www.isle.uiuc.edu/software/lee_vad.m">lee_vad.m.</a>

(Bowon Lee, 2007)
Computation
GMTK Parallel Split GMTK commands into batch jobs for a cluster

<a href="http://mickey.ifp.uiuc.edu/speechWiki/index.php/GMTK_parallel_tools"> Description,</a> <a href="svn://mickey.ifp.uiuc.edu/gmtkScripts/">SVN repository</a>.

(Arthur Kantor, 2008)
HTK Parallel

Split an HTK command into batch jobs for a cluster (Bowon Lee, 2006)
<a href="http://www.ifp.uiuc.edu/~bowonlee/research/cluster/HTK_parallel.htm"> Description</a>; <a href="svn://mickey.ifp.uiuc.edu/HTK_parallel/">SVN repository</a>; <a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HCopy.pl"> HCopy.pl</a>, <a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HVite.pl"> HVite.pl</a>, <a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HERest.pl"> HERest.pl</a>, <a href="http://www.ifp.uiuc.edu/~bowonlee/research/htk-pl/HResults.pl">

HResults.pl</a>.
Data
dtmfsegSegment audio files at DTMF tones

<a href="svn://mickey.ifp.uiuc.edu/dtmfseg/">SVN repository</a>

(Bowon Lee, 2006)
transcription toolsConvert transcription formats

<a href="svn://mickey.ifp.uiuc.edu/transcription_tools/">SVN repository</a> <a href="http://www.isle.uiuc.edu/software/transcription_tools2005May.tgz"> TGZ archive</a>

(Mark Hasegawa-Johnson, 2005)
speechfileformatsRead and write HTK files in matlab

<a href="svn://mickey.ifp.uiuc.edu/speechfileformats/">SVN repository</a> <a href="http://www.isle.uiuc.edu/software/speechfileformats.tgz"> TGZ archive</a>

(Mark Hasegawa-Johnson, 2004)
CTMReditManually and automatically

segment CT and MR image stacks
<a href="http://www.isle.uiuc.edu/pubs/1990s/hasegawa-johnson99embs.pdf"> Description</a>; <a href="svn://mickey.ifp.uiuc.edu/CTMRedit/">SVN repository</a> (Mark Hasegawa-Johnson and Jul Cha, 1999)


Phonetic Transcription Tool

This tool gives the American English phonetic transcription of any string. It uses an HMM model to either generate a most likely phonetic transcription, or if a phonetic transcription is provided, it can perform forced alignment. So, it gives a reasonable pronounciation for out-of-dictionary words, or partially pronounced words.

Phonetic Transcription Tool. You can try out the demo here.

Perl scripts for parallel processing of HTK commands

Perl scripts for parallel processing of four HTK commands (HCopy, HERest, HVite, and HResults) are available.

1. HCopy.pl

2. HERest.pl

3. HVite.pl

4. HResults.pl

Those scripts use the SGE (Sun Grid Engine) for job queuing.

Detailed information about the above Perl scripts can be found here.

Brief introduction about the SGE can be found here


Written by Bowon Lee, 02/24/2006

Retrieved from "http://mickey.ifp.illinois.edu/speechWiki/index.php/Software"
Personal tools