As a Cal Poly computer science thesis project I investigated whether flute tone quality can be objectively quantified in a meaningful way (click on this link for more information about me).
I was perhaps 5 years old when my Uncle was explaining some basics about the flute to my older brother. As soon as my Uncle started playing I was stunned, the entire ambience of the room changed. It was a transcendent moment... how could someone blowing into a little metal tube alter my state of mind and produce such a profound visceral effect?
Forty years later I started learning to play the flute. As you might imagine, being an adult beginner has been a humbling and frustrating experience. I know what I want to sound like, but can't figure out how to get there. Lots of lessons, different teachers, and hours in the practice room... more than 10 years later and still trying.
The experience did inspire my sense of curiosity to understand why I don't sound like I want to. There are ways to measure tone using spectral analysis. Information about this topic can be found on the background page. Once a note has been analyzed, the question is: what correlates with "good" tone quality? This project can was broken down into two parts...
Part I - Build a tool
Musicians use things like tuners and metronomes, so why not use a tool for tone? An application program, Harmonic Analysis Tool (HAT), is under development. Refer to the HAT App page for details.
Part II - What to do with the information from the tool?Assuming a tool can be built to analyze, visualize, and compare tone quality, the question is what is good quality? There is no single ideal tone because "it depends". Different circumstances may call for different qualities. Another factor is everyone has different tastes. A skilled musician will use different timbres as circumstances arise to enhance their musical interpretation and performance. This might be akin to a painter having a palette with a variety of subtle hues to shade his art.
With all this in mind, there is some consensus on standards for tone. For example, a beginning elementary school flutist will probably produce a sound that is less desirable than a flutist playing in one of the major symphony orchestras. The second part of this project attempted to correlate flute harmonic profiles (or harmonic signatures) with qualities of tone. The methodology will be to consult with experts as well as perform surveys of flutists and other musicians.
Papers and Related Webpages
- Using Spectral Analysis To Evaluate Flute Tone Quality (thesis paper)
- Proceedings, Acoustical Society of America (condensed version of the thesis)
- Investigating the Usability of Existing Mobile Training Apps (usability study for a prototype of HAT)
- Spectra and audio clips of the survey samples
- Preliminary survey results
- Spectra and audio clips of Syrinx analysis for 9 professional recordings
- Jean-Pierre Rampal - 7 different Syrinx recordings over the years