Re: Robust method of fundamental frequency estimation.

["Alain de Cheveigne'" <Alain.de.Cheveigne@xxxxxx>]

> Those who are interested can find the YIN link at 
> http://www.auditory.org/postings/2002/26.html.

You can also find the matlab implementation of YIN at: 
http://cognition.ens.fr/Audition/adc/sw/yin.zip.  Caveats: the 
implementation calls routines in C ('mex functions') that you may 
need to recompile for your machine, and the code is hard to read. 
YIN may be the most effective way to use the autocorrelation approach.

For those interested in the wider issues involved in estimating F0 
and relating it to pitch, my review chapters on multiple F0 
estimation 
(http://cognition.ens.fr/Audition/adc/pdf/2006_casabook.pdf) and 
pitch perception models 
(http://cognition.ens.fr/Audition/adc/pdf/2005_pitch_SHAR.pdf) might 
be of use.  The YIN paper is at 
(http://cognition.ens.fr/Audition/adc/pdf/2002_JASA_YIN.pdf).

Alain