[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: f0 frequency difference limen

In terms of musical intervals, 0.2% for partials, 0.1% for the F0
translates into 3.4 and 1.7 cents, respectively, where 100 cents
corresponds to an equal-tempered semitone change of pitch.

I wouldn't think that this result would be completely independent
of F0. I would expect the percentage threshold to increase for
F0s below a certain point. It would also depend on the spectrum.
I.e., spectra with just a few partials in the bass region may
have much larger DLs. But 2 cents is a good practical df to
shoot for in any music synthesis system.


Original message:
>From: "Alain de Cheveigne'" <Alain.de.Cheveigne@xxxxxx>
>Date: Wed, 14 Jan 2009 12:29:12 +0000
>To: AUDITORY@xxxxxxxxxxxxxxx
>Subject: Re: [AUDITORY] f0 frequency difference limen
>Comments: To: andymilne@xxxxxxxxxxxxxx
>According to [Henning, G. B. and Grosberg, S. L. (1968). "Effect of 
>harmonic components on frequency discrimination." J. Acoust. Soc. Am. 
>44: 1386-1389.], for fundamental frequencies below 2 kHz, 
>discrimination is better for complex than pure tones (above 2 kHz it 
>is the same).
>According to the theory of [Goldstein, J. L. (1973). "An optimum 
>processor theory for the central formation of the pitch of complex 
>tones." J. Acoust. Soc. Am. 54: 1496-1516.], F0 discrimination of a 
>complex is at least as good as that of each of its components.
>Both suggest that complex tone limens should be at least as small as 
>pure tone limens if not better. The classic source for those is 
>[Moore, B. C. J. (1973). "Frequency difference limens for 
>short-duration tones." J. Acoust. Soc. Am. 54: 610-619.].  If I 
>remember correctly, pure tone FDLs are as low as 0.2% at 2kHz.
>For a more recent study see [Gockel, H., Moore, B. C. J., Carlyon, R. 
>P. and Plack, C. J. (2007). "Effect of duration on the frequency 
>discrimination of individual partials in a complex tone and on the 
>discrimination of fundamental frequency." J. Acoust. Soc. Am. 121: 
>373-382.].  They report thresholds as low as 0.2% for partials, 0.1% 
>for the complex.