Run Acoustic Model Creation Script

Note: you can download the files using the individual links below, or download the VoxForge tutorial/howto zip file and drag and drop the files to the correct folders.

Download the following Julia scripts to your 'voxforge/bin' folder.

Next, create a new "input_files" folder inside your voxforge directory (so that you have 'voxforge/howto/input_files' ), and download these files into it:

 

Run the acoustic model training script:

cd  voxforge/howto
julia ../bin/trainAM.jl

This will create the following files (located in the howto/acoustic_model folder) that make up your acoustic model:

Note: the hmmdefs file shown here was trained with my voice.  It will not work with your voice.  Your file will have a the same structure as this one, but the statistics will be different.
 

 

Comments

By resstymanuzon - 10/11/2018 - 1 Replies Hello!

By Jame - 7/21/2015 - 1 Replies In another comment they mention that it means a triphone in missing from the sample size. I have the 40 samples from the tutorial and 40 more that include the words Rewind and Restart. The other post have an actual triphone where as this seems to just be a phone. This error accors making hmm13. Sorry if my use of terminology is incorrect.

By angga - 5/31/2014 - 1 Replies ERROR [+2662] FindProtoModel: no proto for k-b+l in hSet

By ajay - 2/22/2014 - 1 Replies error:-

By Michael A - 8/17/2013 - 1 Replies Dear VoxForge,

By thuan - 6/14/2013 - 1 Replies When I run ./HTK_Compile_Model.sh

By thuan - 6/14/2013 When I run ./HTK_Compile_Model.sh

By Mark - 10/8/2012 - 1 Replies I need help for this problem...

By thaparallawadinishant - 1/10/2012 - 2 Replies hello all

By nemozny - 8/30/2011 - 1 Replies If you get "215: let: not found", change in the script the first line to "#!/bin/bash"

By jackinx - 8/29/2011 - 2 Replies Hi, I'm Jack Roh in IIT Delhi.

By Jack Roh - 8/29/2011

By Giovanni - 11/23/2010 - 1 Replies Hi

By swbluto - 9/9/2010 Hello, I tried manually following the steps and I kept getting a ...

By swbluto - 9/9/2010 - 1 Replies The location of the "scripts" folder is slightly confusing that, so I copied and pasted it to both "voxforge/" and "voxforge/auto". After doing that, it resolved the error in step 5 about not finding sample 1 in ../train/wav, and it seems to work correctly until step 13.

By Conrad - 5/22/2010 - 1 Replies All I get is:

By 9yards - 7/21/2009 - 4 Replies I am running HTK_Compile_Model.sh and getting the following error message.

By rahulgupta.87 - 5/4/2009 - 11 Replies

By kmaclean - 1/26/2009 See this entry in trac: Ticket #449 - HowTo error

By colbec - 3/5/2008 - 1 Replies The first time I ran HTK_Compile_Model.sh it stopped at step 5 and complained that it could not find sample1.wav. The message indicated it was looking in a directory one level up, but the Howto calls for the wav files to be stored two levels up in voxforge/train/wav directory. I found that after: mkdir auto/train and mkdir auto/train/wav, copy all sample*.wav files to this new wav directory mkdir auto/train/mfcc then the script ran perfectly to the end. Maybe I missed something, but hope it helps.

By kat_wills - 12/28/2007 - 1 Replies Hi,