Please be Seated

Pleasebe

Seated

The physics of sound:What makes musical

tones different?

Special Lecture for the 2005 Year of Physicsin coordination with the

French National Center for Scientific Researchand the

French Embassy in Washington DC

Special Guests:

Michẻla Castellengo, Research Director, Musical Acoustics Lab, University of ParisHugues Genevoire, Research Engineer, Musical Acoustics Lab, University of ParisCharles Besnainous, Research Engineer, Musical Acoustics Lab, University of ParisJoseph Curtin, stringed instrument makerBenoît Rolland, bow makerSerge de Laubier, musician-researcher

The Overtone Series

Standing waves in a string

The Overtone Series

Standing waves in air columns

The Overtone Series

Standard electronic wave forms

• Sine wave

• Sawtooth wave

• Pulse train

• Triangular wave

• Square wave

Fourier’s Theorem:Any complex wave can be“synthesized” by adding itsharmonics together with the

proper amplitudes and phases.

“Fourier synthesis” and “Fourier analysis”

Fourier Synthesis

Fourier SynthesisSawtooth wave

Fourier SynthesisPulse train wave

Fourier SynthesisTriangular wave

Fourier SynthesisSquare wave

Fourier Analysisor

Spectrum Analysis

Sine Wave Spectrum

Sawtooth Wave Spectrum

Pulse Train Spectrum

Triangular Wave Spectrum

Square Wave Spectrum

Analysis of

Musical Sounds

Recorder Wave and Spectrum

Violin Wave and Spectrum

Crumhorn Wave and Spectrum

Clarinet Wave and Spectrum

Factors in Tone Quality1. Amplitudes of harmonics

2. Attack and decay transients

3. Inharmonicities

4. Formants

5. Vibrato

6. Chorus effect

Vocal Formants

Origin of Vocal Formants(~17.5 cm closed tube)

Frequency:

f1 = 500 Hz

f3 = 1500 Hz

f5 = 2500 Hz

Vocal range:

150-850 Hz

500-2500 Hz

1500-3500 Hz

Mode:

Simple formant model

From Thomas D. Rossing, The Science of Sound, second edition

Vowel formant production

Source: http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html

Vowel Formant Frequencies

From Donald E. Hall, Musical Acoustics, Second Edition

TheSingersFormant

Averaged spectral energy distribution:Light: ordinary speechDark: orchestral accompanimentBrown: Good singer with orchestra

Johan Sundberg: The Acoustics of the Singing Voice; Sci. Amer., March 1977

Sound Spectrograms

Vocal Formant spectra

“OO”

“AH”

“EE”

Vocal Spectrogram of Formants

Voice and Synthesizer “wow”

Matching vocal spectrograms

Kay Elemetrics, Computerized Speech Laboratory

Helium Voice

Singing frequency remains the same

(vibration of vocal folds)

Formant frequencies rise because

She >> Sair

Why?

Vowel formant production

Source: http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html

Speed of Sound in

Helium and Sulfur Hexafluoride

He SF6

M/Mair 1/7 4.6

Vs/S 2.6 0.5

The EndThank you for your attention

We are on the web at

http://www.physics.umd.edu/lecdem/

Animated Gifs compliments of bellsnwhistles.com