Printer Friendly

Sing a song of chaos.

Sing a song of chaos

There seems to be more to the sound of a singing voice than meets the human ear. A new, computer-based technology for automatically converting a singer's tune into musical notation on a computer screen reveals puzzling shifts in sound patterns when a musical note is analyzed segment by segment. The results suggest that the patterns in a sound signal don't always recur in a regular fashion, says Paul E. Dworak of the music school at North Texas State University in Denton. "Rather, a single sound event can be broken down into several modes, with abrupt changes of mode sometimes occurring between two adjacent periods." That type of irregular behavior is characteristic of choatic dynamical systems -- the same kind of behavior that may underlie turbulence in flowing water.

Dworak's intent was to develop a computer system that could evaluate a music student's performance. The computer would first present a melody's musical notes, then display the musical notation for what the student actually sings, allowing the student to identify any wrong notes. Dworak designed and built an electronic circuit that segments the sound wave as it comes in from a microphone. A computer program uses that information to determine a note's pitch.

Instead of looking at the overall shape of waveforms, Dworak's system uses the location of peaks and troughs in the signal to define segments, each one lasting less than a millisecond. "I'm actually looking at the time structure of the sound," he says. "Instead of comparing the shape from cycle to cycle, I'm comparing the time structure from cycle to cycle, which turns out to be more uniform than the shape."

By doing such a cycle-by-cycle analysis, Dworak found that even when the pitch doesn't change in any obvious way, the sound patterns sometimes abruptly shifts into a new configuration after a number of cycles, then suddenly changes again, and so on. Notes played on a guitar and other musical instruments show a similar effect. Whether the effect is audible to the human ear is not yet clear. Dworak is now looking into what causes such perturbations in sound patterns and whether that information can be used to improve the reliability of pitch analyzers. "These structures," he says, "cannot be explained by simple models of sound productions."
COPYRIGHT 1988 Science Service, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:computer analysis of sound of a singing voice
Publication:Science News
Date:May 7, 1988
Previous Article:Warm up with iron.
Next Article:A nickel sponge for cooling air.

Related Articles
Makin' music: songs, rhythm, and creative expression.
Sing your heart out.
Music without Borders.
Children's Pitch Matching, Vocal Range, and Developmentally Appropriate Practice.
The Secrets of Singing.
Music Acquisition of Children in Rural Zimbabwe: A Longitudinal Observation.
Why Catholics shouldn't bother to sing: feeling a little left out during the Communion hymn? It might just be that your parish is dominated by what...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters