This is what I need:
1. I need a simple app that takes in human speech command through a microphone:
2. The app will listen to the following commands:
Up, down, left, right, front, back, zero, one, two..., ten
3. Once the command is spoken and detected, the app displays the command that is read. If not recognized, it displays "unknown"
4. The speech recognition needs to be very precise - to recognize at least 4 commands per second
5. Speaker-dependent speech recognition technology is preferred. That means it's best the speech recognition engine requires speaker training before it is used. I don't think speaker indenpendent engines (like Siri) can do the job.
Do let me know if you have anyone who can do this proof of concept application and let me know the cost. I will need the source code of the software. This can be done on Windows platform.