Abstract
How vocal tract length variation may be incorporated in the Kelly-Lochbaum (1962) speech synthesis model through a dynamic variation of the system sampling rate is described. It has been determined that an eight-point four-term Blackman-Harris (1978) window is optimum for designing the FIR interpolation filter for converting from a varying to a fixed sample rate. >
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have