Development

Browse Source - to view Subversion repository containing VoxForge Speech Audio files, Acoustic Models, and Scripts;
RoadMap - to view the list of outstanding tickets for the current milestone;
New Ticket - to report a bug or request an enhancement.

create Acoustic Models (Linux, Windows) by 'compiling' your speech audio files into a Speaker Dependent Acoustic Model; or
adapt the VoxForge Speaker Independent Acoustic Model to better recognize your voice (Linux, Windows).

Available Projects

Other Dev Pages

    Create Acoustic Models
    Convert Audio to MP3 and Compare Results with Original Wav
    How to Manually Segment an Audio Book (Draft)
    Automated Audio Segmentation Using Forced Alignment (Draft)
    Documentation

Search Dev Comments

VoxForge Dev - Trac

VoxForge Dev Wiki

/uploads/qN/R-/qNR-wXLB0V2rW55uNr7xpQ/hmm.jpg

Create Acoustic Models

/uploads/yz/j0/yzj0ZGufVr95d16QUT2XTQ/applications-office.jpg

Documentation

Recent Dev Tutorial/Howto Threads

(does not include replies to a post)

error by hamid on 9/4/2008 @ 10:38 pm

Error [+6510] LOpen: Unable to open label file downsampled.lab by kmaclean on 8/23/2008 @ 11:33 am

Julius performance by DRF on 8/8/2008 @ 8:28 pm

Makefile Errors? by Ramen on 8/8/2008 @ 5:23 pm

problem with the HERest by imene on 8/8/2008 @ 5:10 pm

Training acoustic model by DRF on 8/3/2008 @ 1:50 am

problem with 1st pass - global adaptation by DRF on 7/21/2008 @ 5:09 pm

Languages

Deutsch
English
Español
Hebrew
Italiano
Nederlands
Português
Русский

Comments

Click the 'Add' link to add a comment to this page; click the 'Read More' link to view replies to a posted comment.

Add • Search

Automated word insertion in alphabetical order

By Sergey - 4/20/2008 - 2 Replies Hello,
Is there a way to add missing words to the VoxForgeDict in alphabetical order automatically?
If no, I will try to write a script.
Thanks,
--Sergey

Missing Words: Semi-automated approach

By Sergey - 4/20/2008 - 3 Replies Hello Ken,
I was trying to segment a real audio book now, and as expected, it has missing words. Step 2 of the tutorial "Add Missing Words: Semi-automated approach using Festival" suggests running MissingWOrdsCleanup.pl script. When I run it, I get an error:

/a/proj/voxforge$ ./missingwordscleanup.pl
festival -b MissingWordsPhonemes.scm
SIOD ERROR: not a symbol or string : (quote S)
closing a file left open: temp
closing a file left open: MissingWords
closing a file left open: MissingWordsPhonemes.scm
system festival -b MissingWordsPhonemes.scm failed: 65280 at ./missingwordscleanup.pl line 51.
/a/proj/voxforge$

and it creates an empty file MissingWords_out.
What might be wrong ?
Thanks,
--Sergey

Automated Audio Segmentation accuracy

By Sergey S. - 4/17/2008 - 11 Replies

Hi Ken,
On 32bit version of Ubuntu HTK compiled without a problem. Then I followed the tutorial "Automated Audio Segmentation Using Forced Alignment" without a hitch up to step 6. In Audacity I see a considerable misalignement between audio and labels. For example in the phrase "Chapter Four, Part One" the audio corresponding to label "part" contains sounds "part one", and the audio corresponding to label "one" contains only silence. Is it a normal misalignement or am I doing something wrong?
Thanks a lot,
--Sergey


Username	Password