[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: F0 estimation software
I suggest you use SWIPE'  to estimate pitch and its companion
TS2means  to distinguish between pitched and unpitched regions.
Matlab code for both of them can be found at
http://www.cise.ufl.edu/~acamacho/publications. The advantages of the
use of SWIPE' vs. autocorrelation, SHS (both as implemented in
Praat), YIN, TEMPO, and others in terms of gross error rates are
documented in .
 Camacho, A., Harris, J. G. (2008). “A sawtooth waveform inspired
pitch estimator for speech and music,” Journal of the Acoustical
Society of America, vol. 124, pp. 1638-1652.
 Camacho, A. (2008). “Detection of Pitched/Unpitched Sound Using
Pitch Strength Clustering,” Proceedings of the Ninth International
Conference on Music Information Retrieval, Philadelphia, pp. 533-537.
On Mon, Jun 7, 2010 at 10:54 PM, Federman, Jeremy
> Hi all -
> We have a graduate student here who is currently investigating the role of
> motor cognition in singing. Specifically, she is recording professional
> singers and non-singers singing familiar tunes in their preferred key
> during masked and unmasked conditions. That is, for some trials the
> participants can hear themselves singing and for others they cannot.
> The purpose of this post is to ask for your help and advice regarding
> software we might potentially use (matlab or other) to automate the
> process of estimating the fundamental frequency (also known as pitch
> detection or F0 estimation) of periodic segments contained in the recorded
> waveforms. If there are known advantages and disadvantages to various
> options, please feel free to include that information as well.
> Thannks in advance for any and all assistance and guidance on this topic!
> With kind regards,
> Jeremy Federman, PhD-ABD, CCC-A
> Vanderbilt University Medical Center
> Department of Hearing and Speech
> 1215 21st Avenue South, Room 8310
> Medical Center East, South Tower
> Nashville, TN 37232-8242
> Email: jeremy.federman@xxxxxxxxxxxxxx