Re: pitch tracking (at)

Subject: Re: pitch tracking
From:    at <lipscombUTDALLAS.EDU>
Date:    Tue, 1 Mar 1994 16:37:36 -0600

I am extremely interested in the issue of pitch tracking addressed recently by Charles Tassoni--specifically, tracking a monophonic vocal or instrumental melody (polyphony opens a whole new can of worms!) I am in the process of creating a Windows-based program that will process a WAV file and produce a graphic output representing frequency (vertically) by time (horizontally) in relation to a piano keyboard located on the leftmost portion of the screen. I have studied the algorithms suggested by Rabiner, et al. My question is this: are there relatively reliable algorithms that are not as computationally intensive as the Rabiner model? I am attempting to create a program that will be useful for illustrative purposes on a typical PC (i.e. 486 processor (at) 33 MHz) without requiring 10 minutes of computation to calculate the results of a 15 second melody. Any ideas? Thanks in advance! Scott Lipscomb

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University