Timbre-streams (PS45000 )

Subject: Timbre-streams
From:    PS45000  <PS45(at)MUSICA.MCGILL.CA>
Date:    Wed, 12 May 1993 20:25:34 EDT

In response to Pierre Divenyi's response to Greg Sandell (relevant parts excerpted below): > Subject: Re: streaming of melodic patterns with high level cues? > > Greg, > > Although I know of no experimental work in this area > (other than Dave Wessel's classic two-timbre streaming > demonstration), the music literature is replete > with timbre/loudness streaming. > > A timbre streaming example of choice is the much-cited > Bach Prelude from the E-major Violin Partita: In > Measure 3 the weak-beat high E's and the strong-beat > lower diatonic motive are segregated not only > as a result of pitch streaming, but also because > of timbre differences due to up-down bowing. > (These timbre differences are very easy to hear.) > More elaborate bowing-bound two-timbre games can > be found at several spots in the Bartok Violin Solosonata > (to stick to that instrument). > > If in doubt, just keep asking those good dead people, they > have tons of examples indicating that they are always ready > to answer any question of this sort that may hit you... > > All best wishes, > Pierre Divenyi _______________________________________________________________ Dear Pierre, Greg , et al, how about catching up on the work of some "good alive people" as well? Experimental work on segregation by timbre has been "streaming" in steadily (before and beyond Wessel's "demo" of 1979). van Noorden did some preliminary studies on streaming based on spectral similarities. (See his 1975 dissertation on Temporal Coherence - available through IPO, Holland, perhaps). The wonderful 1979 review by Steve McAdams and Al Bregman - "Hearing Musical Streams" published in Computer Music Journal (vol. 3, pp. 26-43) also mentions some instances of streaming by timbre. I did a study extending the ideas of van Noorden and Wessel, (F0 streaming vs. spectral content streaming) which was published in a 1987 issue of JASA (vol. 82, pp. 886-899). Al's book on scene analysis (MIT press, 1990) also mentions some other timbre-streaming studies done at McGill. The 1991 paper by Hartmann and Johnson (Music Perc. vol. 9(2), pp. 155-184) explored the influence of many types of stimulus changes, (many of which would be construed as timbre changes) on segregation of interleaved melodies. And to end this chronological review, note that the recent (1993) dissertation of Paul Iverson at Cornell University focussed on this very topic (Auditory stream segregation based on timbre). He is now a post-doc at the U.Washington Speech and Hearing Dept. in Seattle. Contact Pat Kuhl's lab to locate him. My own work on this topic also continues. Some aspects of the spectral vs. temporal cue controversy will be discussed in a paper to be presented next week in Ottawa (see Pgm. p. 2363 if you're interested). The "good dead people" are always providing questions and answers as they continue to de-compose! And so shall we. Sincerely, Punita G. Singh

This message came from the mail archive
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University