Hello folks,
I am a bit new to the DSK 5510 processor.I have been given the task of differentiating the 3 phonemes 'aa','ee' and 'oo' from their FFT plots. I speak into the microphone these voices and notice the change in FFT. I see that there isn't much change in the FFT of 'oo' and 'ee'. Hence,i am finding it tough to come up with a suitable algorithm to distinguish between these phonemes.
One problem that makes my task difficult is that as i run my program there is already some disturbance in the FFT graph even when I am speaking nothing into the microphone.Is there any way I can reduce this noise.
Does my sampling rate in my Program make a difference.
I would really appreciate any kind of help.
Thanking You
Uday khandelwal