How to build a QT application for speech recognition???

**Gokulnathvc** · 11th March 2011, 12:06

Qt application, which should get the speech as input. For instance, user control, a Mp3 player, if user says "start", it should start playing and if the user says "stop", it has to stop,... pls help me..

**squidge** · 11th March 2011, 12:58

How are you currently receiving the microphone input?

Which speech recognition API as you using?

**Gokulnathvc** · 11th March 2011, 13:06

No, Trying to build in QT itself. No idea about how to begin in QT...

**SixDegrees** · 11th March 2011, 17:16

You've bitten off far too much, then. If you don't know anything about speech recognition, start here: it's one of the most difficult problems in computer science at the moment, and is far from solution.

You'd be far better off researching COTS solutions that can be adapted to your particular needs than attempting to invent them from pure scratch.

**Aladin** · 14th March 2011, 16:02

You can use pocketsphinx on linux,it's a useful library.

**freely** · 15th March 2011, 15:17

I have idea:
a) Qt has all high level wrappers for underluying audio system ( hw and drivers ) - so you access you mic. using QMultimedia module or if at mobile so refer to QtMobility
b) You should "parse" your samples gathered from a) to some buffer every time interval and when you find needed samples combinations ( by some pattern, threshold ) you call Qt slot or event and there is a code that treats what to do on pattern.
Sir I can make for you the application I already knows how to build it ( it is not so sofisticated), or I can help you to build it if you want to cooperate with me contact me personally: pmazniker@gmail.com
Pavel