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

Re: Detection of harmonics and rhythmic structure

>Date:    Fri, 12 May 2000 13:33:29 +0200
>From:    =?X-UNKNOWN?Q?T=F3th_L=E1szl=F3?= <tothl@INF.U-SZEGED.HU>
>Subject: Re: Detection of harmonics and rhythmic structure
>On Fri, 12 May 2000, Neil Todd wrote:
>> compute the modulation power spectrum (see refs below).
>> the most recent version of this has the following components
>> (i)   outer-middle ear,         standard table
>> (ii)  gammatone filter bank
>> (iii) hair-cell
>> (iv)  low-pass                  40 Hz
>> (v)   onset-cells
>> (vi)  modulation band-pass filter bank,         approx. 0.5 - 16 Hz
>The other components are clear, but could you point me towards
>a description of an onset-cell model (or tell me which of your references
>contains the exact description of the model)?
>               Laszlo Toth
>        Hungarian Academy of Sciences         *
>  Research Group on Artificial Intelligence   *   "Failure only begins
>     e-mail: tothl@inf.u-szeged.hu            *    when you stop trying"
>     http://www.inf.u-szeged.hu/~tothl        *


Here are some more details.

The ear model is that developed by Martin Cooke and Guy Brown (Cooke, 1992; Brown and Cooke, 1994)
i.e the gammatone filter-bank is spaced on the ERB rate scale (Glasberg and Moore, 1992) and
combined with the Meddis (1988) hair-cell. The  onset-cell is described in (Brown and Cooke, 1994).
The modulation filters  are such that the filter-bank is constant-Q, i.e. constant bandwidth.   When
convolved with the summed onsets the output of this stage is represented by the modulus and phase of
the filter-bank response.  The details of the filter (2nd order band-pass) is described in Todd
(1994). By happy coincidence it is the same filter (although in a different configuration) used by
Torsten Dau at Oldenburg (1996) to model modulation detection, which supports my argument that the
same mechanism mediates both modulation detection/discrimination and time-interval/tempo
discrimination (Todd, 1996).

Brown, G.J. and Cooke, M. (1994)  Perceptual grouping of musical sounds: A computational model
J. New Music Research. 23(2), 107-132.

Cooke, M. (1992) Modelling Auditory Processing and Organisation. CUP, Cambridge.

Dau, T. (1996) Modelling Auditory Processing of Amplitude Modulation. Doctoral Thesis,
University of Oldenberg.

Glasberg, B. and Moore, B. (1990) Derivation of auditory filter shapes from notched-noise data.
Hear. Res. 47, 103-138.

Meddis, R. (1988) Simulation of  auditory-neural transduction: Further Studies.
JASA 83(3), 1056-1063.

Todd, N.P McAngus (1994) The auditory primal sketch: A multi-scale model of rhythmic grouping.
Journal of new Music Research, 23(1), 25-70.

Todd, N.P McAngus (1996) Time discrimination and AM detection.
JASA 100(4), pt2, 2752.