Speech Recognition Engines

Flat
Helped me run program Language models: Java Speech Grammar Format.
User: DaoDiep
Date: 5/2/2013 9:44 pm
Views: 2564
Rating: 2

I read website: http://cmusphinx.sourceforge.net/wiki/

I using Sphinx speech recognition in Linux with model language statistics and The file structure for the database is

etc

your_db.dic - Phonetic dictionary

your_db.phone - Phoneset file

your_db.lm.DMP - Language model

your_db.filler - List of fillers

your_db_train.fileids - List of files for training

your_db_train.transcription - Transcription for training

your_db_test.fileids - List of files for testing

your_db_test.transcription - Transcription for testing

Then, I Run Program with 2 command line:

1. sphinxtrain - t vn setup

2. sphinxtrain run

but program of me have been rate errors word many

My prorgam is speech recognition the digits word 0-9. I use Centos

Then, I use Building language models: Java Speech Grammar Format. 

But, How I do run program ?

when i run : sphinxtrain - t vn setup create sphinxtrain.cfg and automatic request :your_db.lm.DMP - Language model

I train model language statistics create file .lm.DMP but with jsgf how do create it?

I don't know run program sphinxtrain.

Do you help me.

Thanks you so much

--- (Edited on 5/2/2013 9:44 pm [GMT-0500] by DaoDiep) ---

PreviousNext