Overtone singing needs an extremely narrow, movable vocal-tract resonance over a single harmonic.

Evidence

To achieve this, a very narrow band resonance is required from the vocal tract that can be moved in center frequency to be placed over a desired harmonic, in the same way that the formants modify the amplitudes of the harmonics of the vocal sound source in singing. A narrow-band spectrogram with a spectrum at the instant marked with the vertical line for a snippet of overtone singing is shown in Figure 4.38. The underlying drone note is F#2 with an f0 of 93 Hz. Since the underlying vowel is fixed, the overtones can be seen as any harmonics whose level (darkness of marking) varies, and in this case, it is approximately the twentieth to the twenty-fifth harmonics that are employed. The spectrum indicates how narrow the resonance is that enhances the overtone compared with the vowel formants at that instant. — David M. Howard, Jamie Angus, Acoustics and Psychoacoustics, 1661

Categorization

  • XREF: Connects to the earlier formant and source-filter material — same mechanism, far narrower bandwidth. — connects to: Overtone singing selects harmonics with sharp vocal-tract resonances to create melodies — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 632; Pulling the vocal tract resonance down bends notes but is self-limiting. — Eric J. Heller, Why You Hear What You Hear- An…, 1626; High-pitched soprano singing is unintelligible because harmonics are too sparse for formants — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 631 (+7 more)
  • SHARE: For the one person into vocal technique or spectrograms: the drone is F#2 at 93 Hz and the tune rides harmonics twenty through twenty-five.

Broader Connections