Reply by phitenk August 4, 20082008-08-04
Hi,

i need snr comparision of lpc-10 and voice-excited LPC (VELP) to show
that velp is better as compared to lpc.
and why in velp synthesizer, we must use DCT+noise and don't use pitch
computation?
by computing the pitch so we can determine whether the input speech
frame is voiced or unvoiced, if we don't use pitch in synthesizer, is
DCT can determine input speech voiced or unvoiced?!
can any one help.
thanks.

regards,
phitenk