Sound Fun
Listening for Defects
Listening for Defects: Wavelet-Based
Acoustical Signal Processing in Japan
SIAM News, Vol. 29, No. 2, March 1996
- Listening for Defects
Wavelet Records
An offbeat link for an offbeat page!
- Wavelet Records
The Wavelet Transform as a Musical Score
- As a Musical Score
Whale Calls
Mark Fischer has been analyzing whale calls in wavelet space, producing beautiful wavelet transforms that I think could stand in as a new form of art. For example, see this
- Blue whale "A" call (165 kBytes),
which was transformed with a Biorthogonal 3.1 wavelet. (The vertical axis is not linear, so the bottom is a little stretched.) You can see a gallery of transformed whale sounds of at interspecies.com, watch and hear different species on Google Video, and read a nice New York Times article about his work. Finally, don't miss his fine collection of MP3 sounds from whales, dolphins and assorted sea life, too!
Tonebursts
Victor Wickerhauser has suggested that sound synthesis is a natural use of wavelets. If one wishes to approximate the sound of a musical instrument, the notes can be decomposed into its wavelet packet coefficients. Reproducing the note would then require reloading those coefficients into a wavelet packet generator and playing back the result. Transient characteristics such as attack and decay can be controlled separately (for example, with envelope generators), or by using longer wave packets and encoding those properties, as well, into each note.
Wickerhauser has done just this. He has created combinations of wave packets that produces especially interesting sounds. In Fall 1994, he kindly gave me one these wavelet tonebursts to use (many thanks!). The sound of this particular toneburst is haunting. I found that I couldn't get the sound out of my head for days after hearing it! If your Web browser and/or WWW helper application supports .au sound files, then hear this wavelet toneburst sound for yourself.
- Press
to hear a wavelet toneburst (110 kB).
Wavelets in Electro-Acoustic Music
Corey Cheng's earlier work included a fine master's thesis which, among many things, described analogies between fugues and wavelets! With Hofstadter and Cheng, my wavelet introductory presentations gained new ear-grabbing hooks to interest people learning about wavelets for the first time.
Audio Analysis using the Discrete Wavelet Transform
This is a readable paper I found somewhere that I liked a lot.
The abstract reads:
The Discrete Wavelet Transform (DWT) is a transformation that can be used to analyze the temporal and spectral properties of non-stationary signals like audio. In this paper we describe some applications of the DWT to the problem of extracting information from non-speech audio. More specifically automatic classification of various types of audio using the DWT is described and compared with other traditional feature extractors proposed in the literature. In addition, a technique for detecting the beat attributes of music is presented. Both synthetic and real world stimuli were used to evaluate the performance of the beat detection algorithm.
Violins
In Summer 2002, I was bitten with the curiosity bug of the sound of Cremona violins after reading a Scientific American web contest to try to tell the difference between a Nagyvary violin and a Stradivarius violin. I had some difficulty with distinguishing the pieces. I used a variety of techniques including performing a wavelet analysis and an FFT on the given violin samples, and a wavelet analysis of a sound sample of Nigel Kennedy playing a Stradivarius.
Some results of discrete wavelet transforms of violin samples:
The idea of analyzing these special Cremona-made violins in frequency space is not new- it's been done for more than fifty years, however the results are today still controversial. From my brief foray, I know that my samples were too short and too small, and now I am exploring further with Mr. Nagyvary the idea of distinguishing these sounds in wavelet space. I doubt that I am the first to use wavelet transforms to look at sounds from Cremona-made violins, but so far I've not seen other wavelet analysis performed. Therefore, if you know of people and papers describing wavelet analysis on violin sound samples, please drop me a note.
Kirk in Wavelets
Breaking down sound samples using wavelets tickles my funny bone and sometimes I share my sound-fun when I introduce people to wavelets.
If you've not heard Star Trek's Captain Kirk's true voice because your country's television dubs his voice with something completely different, then here I offer you the next best thing: Kirk -- in wavelets.
For your own fun and amusement, I present to you a comparison of Kirk's "thank you" with my "thank you". Then, after analyzing a starship captain's coefficients with an astronomer's coefficients, you can form your own theory about whether this astronomer (moi), was staring at a few too many stars when she wrote this wavelets page.
Wavelet Page
|
© Copyright Amara Graps, 2002-6. |