Comments on: Basics of Java Speech Grammar Format ( JSGF )

By: Akki

Akki — Fri, 16 Mar 2012 05:48:37 +0000

Hi Puneet, i need to know how to implement dictation grammar(i mean adding words to grammar file dymnamically[which are not present in grammar file]),Can you please post source code for that)
with framework which you are going to use for thateg:sphinx4,freetts..

By: Amit

Amit — Wed, 04 Jan 2012 13:13:26 +0000

For Speech Recognition I have decided to use sphinx4, I am trying to run the demo Transcriber.jar provided with sphinx. That doesn’t work when I give some other file as input.

How do I proceed ? Where do I get the language model for US English that can be used with Sphinx4 ?

Any blog / tutorial will be helpful,

By: Tajinderpal singh

Tajinderpal singh — Wed, 09 Nov 2011 17:49:51 +0000

Hello,puneet i have created a media player in java .I want to use voice recognition using sphinx to stop and play the songs using the voice .I have seen you video you have used eclipse but i am using command prompt which does’nt give same result .iam also using apache ant to build sphinx pgm through ant command but when i run the pgm i am getting the following error
Buildfile:build.xml does’nt exist.
can you tell me the procedure of running the voice recognition pgm through command prompt.I will be greatly thankful to you please reply fast .it’s very urgent.

By: Another Title

Another Title — Sat, 08 Oct 2011 17:39:21 +0000

tv online…

http://www.tv-online-live.com…

By: Reshma

Reshma — Mon, 26 Sep 2011 05:34:44 +0000

u mean v will be able to recognize 1000 words???.. then wat will be the grammar… do v need explicitly specify all the 1000 tokens that is present in the dictionary??.. please answer this…

By: Puneet Kalra

Puneet Kalra — Fri, 01 Jul 2011 11:53:31 +0000

Hey Varun,

I’m afraid there’s no such way to do that with Sphinx4. But yea, if your speech will be limited to few thousand words then of course yes!

You will need to work on Acoustic Model, Language Model and Sphinx config as well.

Regards,
Puneet Kalra

By: Varun Vijaywargi

Varun Vijaywargi — Fri, 01 Jul 2011 09:38:35 +0000

Hi Puneet,
First of all, great work by you.
I was able to run the hello world application. However, now I am trying to develop a more complicated application and it should use the runtime dictionary (i.e. it should recognize whatever I say)
Do you have any idea of how to do it?

Thanks in Advance,
Varun Vijaywargi

By: Abhinay saxena

Abhinay saxena — Fri, 13 May 2011 04:08:18 +0000

hi,
i am an engg student i have created speech recog for fedora I have to increase my lib and grammar.
so i want to know does it require any wave file for recognition or just this grammar can do?????

By: arun

arun — Mon, 25 Apr 2011 06:00:20 +0000

I am facing problem in allocating heap in eclipse and in you tutorial resolution is not good so not understood allocation o heap 2nd video tutorial can you explain more

By: shashank

shashank — Sat, 12 Feb 2011 08:56:39 +0000

this blog is really proving worth to every one who wanna work with speech recognition thanx for such hard work and providing it here