Click here to Skip to main content
15,887,214 members
Please Sign up or sign in to vote.
0.00/5 (No votes)
See more:
Hello,

I want to build an application like the one in the old cellphones, where you could record someone's name, and then call that person by saying it (for example, while you were driving).

I tried microsoft's speech SDK, but it didn't even come close to understanding what I'm trying to say (I have bad accent).

Generally I feel that using a general speech recognition is a little overkill for what I'm trying to achive. Basically I'm trying to solve a very narrowed version of SR.
I tried playing with audio signals comparison, but I have no knowledge of the subject.

Anyone knows if there is a library that does what I need, or some direction to how they did it in the old cellphones?

Thanks alot,
Elad
Posted

1 solution

Perhaps you tried to big a thing when you played with Microsoft Speech Recognition: using its full vocabulary.
Create a small vocabulary - containing only the names you want to recognize, and their pronounciation. You could start with recognizing simple numbers (one, two, three,...) before you try names. Then results ought to be better.
 
Share this answer
 
Comments
eladweiss06 6-Aug-12 9:05am    
No. I tested it with one word in the vocabulary...

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)



CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900