I have a simple 'working' voice recognition add-in for Windows Vista Media Center, made in Visual C. It is using Vistas built-in voice recognition in English/US to work.
I has a bug, and a couple of missing features.
Basically...You can say 'Trigger word' + Command, and it will react to that...mostly
Fx "Media Center --- Go to Channel 6"
Or "Media Center --- Go to My Music" etc...
The problem, I think, is that with some of the commands, it seems to skip the Trigger word, and just react to the command, making it very sensitive to voices from the TV, making it changing channels, etc. on a random basis when the volume is high.
Missing feature 1.
It can react to the spoken numbers, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0. So if you want channel 14, you will have to say
"Media center --- Channel 9" & then "Media Center --- Channel Up" 5 times.
So I want it to react to the numbers, 10-20 as well.
Missing feature 2.
If you go to "My Music" and have a lot of music files in that folder, and you want to listen to a song in the middle, you will have to say "Media Center -- Go Right"
many times, untill you reach the file you want to listen to. So I want it to be able to go to the files starting with: A, B, C, D, E, F, Etc.
So basically..you must have Vista with with built-in voice recognition in English/US, a microphone, and knowledge about Media Center, the Voice recognition Sapi Vista has, and probably the 'codes' a MCE remote control is using.
I would like to have the project finished today if possible.