Re: Modelling of musical instrument spectra

[beaucham <beaucham@xxxxxxxxxxxxxxxxxxxxxx>]

Dave,

There are quite a lot of papers which give spectra of instruments but
very few that give phase information or spectral models. Here is an
old one that does address the spectral analysis of violin tones giving
both amplitude and phase information:

Beauchamp, J. W. (1974). "Time-variant spectra of violin tones", J. Acoust.
Soc. Am. 56(3), pp. 995-1004.

There have been a few attempts at spectral-envelope-based models:

Luce, D. and Clark, M. (1967). "Physical Correlates of Brass-Instrument
Tones", J. Acoust. Soc. Am., Vol. 42(6), 1232-1243.

Luce, D. A. (1975). "Dynamic Spectrum Changes of Orchestral Instruments",
J. Audio Eng. Soc. 23(7), 565-568.

Horner, A. and Beauchamp, J. (1995). "Synthesis of Trumpet Tones Using a
Wavetable and Dynamic Filter", J. Audio Eng. Soc. 43(10), 799-812.

Beauchamp, J. W. and Horner, A. (1995). "Wavetable Interpolation Synthesis
Based on Time-Variant Spectral Analysis of Musical Sounds", Audio Eng. Soc.
Preprint No. 3960, pp. 1-17.

Also, David Luce's doctoral thesis, "Physical Correlates of Non-Percussive
Musical Instrument Tones", M.I.T. Physics Dept., 1963, has a wealth of
data to look at. The spectral envelope stuff is summarized in his 1967 and
1975 papers.

Some early copies of the Computer Music Journal have graphs of harmonic
amplitudes vs. time along with actual numeric data for various instruments
at selected pitches. A key word in the title of the papers is "Lexicon".

Jim Beauchamp
Univ. of Illinois at Urbana-Champaign
jwbeauch@xxxxxxxx

Original message:
>From: David Gunawan <d.gunawan@xxxxxxxxxxxxxxxxxxx>
>Date: Mon, 17 Jan 2005 14:57:22 +1100
>To: AUDITORY@xxxxxxxxxxxxxxx
>Subject: Modelling of musical instrument spectra
>
>Dear list,
>
>I am looking for spectral models of various musical instruments, preferably
>models related to how harmonic amplitudes and phases move over time. Any key
>publications, books, or other references would be greatly appreciated.
>
>Thanks,
>Dave