cutting audio files based on a keyword

Question

Hi,

I have a audio file(s) that need to be cut and broken up into multiple audio files based on a keyword. For example, lets say the keyword is "CUT"

if we had an wav file called "original.wav" with the following audio, 
"Hello , is this CUT the time is CUT My name is CUT The balance is"

and the token audio cut.wav which contains the audio "CUT"

So original.wav, and cut.wav are feed into a program

and the output is

file1.wav which contains audio "Hello, is this"
file2.wav which contains audio "the time is"
file3.wav which contains audio "My name is"
file4.wav which contains audio "The balance is"

I have no experience in audio programming at all, what libraries would I need and how would I go about this.

Thanks

Answer 1

If you have no experience in this then you'd be better off cutting your audio files manually using Audacity.

It sounds like you're trying to save yourself the effort of cutting your recordings manually, but speech recognition is a very complex topic. You'll spend orders of magnitude more time implementing/integrating your speech recognition engine and training the models than you would spend redesigning your application with the recordings cut 'by hand'.

If you must, you can look at the Microsoft Speech API. The Open Directory also has several links.