[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Fwd: ERBs and resolvability]
Dear List,
here is a summary of comments I have received concerning the question of
resolved and unresolved harmonics.
Best regards,
Annemarie
--- Begin Message ---
Dear Annmarie,
I have proposed that partials within (equal-amplitude) complex tones can be
resolved when they are separated from neighbouring partials by 1.25 ERBs; see
Moore, B. C. J. and Ohgushi, K. (1993) Audibility of partials in inharmonic
complex tones. J. Acoust. Soc. Am. 93, 452-461.
Moore, B. C. J. (1997) An Introduction to the Psychology of Hearing, 4th
Ed., Academic, San Diego.
Using this assumption, it is easy to work out, for a given f0, the harmonic
that will first be unresolved. You can download from the website given
below a simple program that will give you the centre frequency for a given
ERB width.
Example:
Assume f0 = 100 Hz. The harmonics will be at the threshold of
resolvability when the ERB is 100/1.25 Hz, i.e. 80 Hz.
To use the program type freq2erb -e (return)
and enter 80 Hz. The program will give a centre frequency of 513.1 Hz.
Thus, the 5th harmonic will just be resolved, but the sixth will not.
Best wishes,
Brian
The center frequency at which the ERB is 80 Hz is
Brian C. J. Moore, Ph.D.
Professor of Auditory Perception,
Department of Experimental Psychology,
University of Cambridge,
Downing Street,
Cambridge CB2 3EB,
England
Tel. + 44 1223 333574
Fax. + 44 1223 333564
http://hearing.psychol.cam.ac.uk/
--- End Message ---
--- Begin Message ---
annemarie,
look in shackleton & carlyon (1994), jasa vol 95 p 3529-3540. The limit
depends on tehfrequency region into which the complex is filtered. A simple
rule is in he number of harmonics in teh 10-dB down bandwidth of an
auditory filter centered on the complex
<2=resolved
2 to 3.25 ambiguous
>3.25 unresolved
10-dB down BW i=44.46(4.37F+1) where F is in kHz
bob
At 13:09 29/09/00 +0100, you wrote:
>Dear List,
>
>I am looking for an adequate method to determine the frequency limit
>separating resolved from unresolved harmonics in a complex tone, when f0
>is given.
>I am interested in empirical data, simple algorihms based on ERBs, as
>well as in program listings (Fortran, C, Matlab etc.).
>
>
>Thanks in advance!
>
>Annemarie Seither-Preisler
>
>
Dr. Bob Carlyon
MRC Cognition and Brain Sciences Unit
15 Chaucer Rd.
Cambridge CB2 2EF
England
Phone: (44) 1223 355294 ext 720
Fax: (44) 1223 359062
NOTE NEW EMAIL ADDRESS:
bob.carlyon@mrc-cbu.cam.ac.uk
--- End Message ---
--- Begin Message ---
OK here's a little start ... I have attatched simple scripts which should
work in 'octave' http://www.che.wisc.edu/octave (perhaps Matlab too -
don't know though - if you run Matlab please inform your parents)
The scripts will find ERBs from Frequencies and vica versa.
Once in the ERB domain harmonics are equidistant in range (in theory any
way). But I'm sure someone else will want to expand on this point ....
There are some papers by Greenwood for example which will tell you that
what is reguarded a harmonic is different for different mammals ...
Matt
On Fri, 29 Sep 2000, Annemarie Seither-Preisler wrote:
> Dear List,
>
> I am looking for an adequate method to determine the frequency limit
> separating resolved from unresolved harmonics in a complex tone, when f0
> is given.
> I am interested in empirical data, simple algorihms based on ERBs, as
> well as in program listings (Fortran, C, Matlab etc.).
>
>
> Thanks in advance!
>
> Annemarie Seither-Preisler
>
function freq=erb2freq(erb)
%# ERB2FREQ - Finds the freq. in Hz for an erb.
%# references
%#[1] ``A Model for the Prediction of Thresholds, Loudness, and Partial
%# Loudness'' Moore B.C.J., Glasberg B.R. and Baer T., Journal of the
%# Audio Engineering Society, vol. 45, no. 4, April 1997, pp.224-40.
c1=24.673;
c2=4.368;
c3=(2302.6/(c1 * c2));
freq=1000.0 * (10.^(erb/c3) - 1.0) / c2;
return;
function erb=freq2erb(freq)
%# FREQ2ERB - Finds the erb for a freq. in Hz
%# references
%#[1] ``A Model for the Prediction of Thresholds, Loudness, and Partial
%# Loudness'' Moore B.C.J., Glasberg B.R. and Baer T., Journal of the
%# Audio Engineering Society, vol. 45, no. 4, April 1997, pp.224-40.
c1=24.673;
c2=4.368;
c3=(2302.6/(c1 * c2));
erb=(c3*log10((c2 * freq/1000.0) + 1.0));
return;
--- End Message ---
--- Begin Message ---
On Sat, 30 Sep 2000, flatmax wrote:
> OK here's a little start ... I have attatched simple scripts which should
> work in 'octave' http://www.che.wisc.edu/octave (perhaps Matlab too -
> don't know though - if you run Matlab please inform your parents)
>
> The scripts will find ERBs from Frequencies and vica versa.
>
> Once in the ERB domain harmonics are equidistant in range (in theory any
> way). But I'm sure someone else will want to expand on this point ....
>
> There are some papers by Greenwood for example which will tell you that
> what is reguarded a harmonic is different for different mammals ...
Not really. Harmonics are the same in any language (or animal).
But see:
Greenwood, D.D. (1990) A cochlear frequency-position function for several species - 29 years later. J. Acoust. Soc. Am. 87, 2592-2605.
Greenwood, D.D. (1991) "Critical bandwidth and consonance in relation to cochlear frequency- position coordinates," Hearing Res. 54, 164-208.
Greenwood, D.D. (1996) "Comparing octaves, frequency ranges, and cochlear-map curvature across species, Hearing Reseach, 94, 157-162.
The second and third papers in particular contain what Matt has in mind. In the second, there are replots of Mayer, Plomp and Steeneken, and others on consonant dyads and Plomp (Ear as Frequency Analyser - considering harmonic sequences), etc. This seems relevant to what you are driving at.
Inquire further as needed.
Donald Greenwood
> On Fri, 29 Sep 2000, Annemarie Seither-Preisler wrote:
>
> > Dear List,
> >
> > I am looking for an adequate method to determine the frequency limit
> > separating resolved from unresolved harmonics in a complex tone, when f0
> > is given.
> > I am interested in empirical data, simple algorihms based on ERBs, as
> > well as in program listings (Fortran, C, Matlab etc.).
> >
> >
> > Thanks in advance!
> >
> > Annemarie Seither-Preisler
> >
>
> ``
--- End Message ---
--- Begin Message ---
Dear Anne-Marie,
Would you please share with the rest of us all the
helpful comments you have received? Apparently, you
got them personally, rather than through the list.
Thank you and best regards,
Pierre Divenyi
=====
__________________________________________________
Do You Yahoo!?
Yahoo! Photos - 35mm Quality Prints, Now Get 15 Free!
http://photos.yahoo.com/
--- End Message ---