Abstract

This paper describes a new algorithm to identify voiced sections in a speech waveform and determine their pitch contours. The new pitch determination algorithm (PDA) uses variable frames and thus overcomes some common problems associated with fixed frame PDAs. It is adapted from an existing maximum likelihood PDA by retaining its strength and removing some weakness in the original formulation. The new PDA was tested on sixteen sentences from TIMIT database and its effectiveness is supported by the test results.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call