High Voltage Forum
General electronics => Electronic Circuits => Topic started by: Physikfan on January 08, 2019, 08:44:16 PM
-
The acoustic spectra of sung vowels of several persons were recorded with a condenser microphone and the FFT-Analyzer HP3561A:
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2Fgalerie.experimentierkasten-board.de%2Fdata%2Fmedia%2F201%2FIreneIngaArthur3F800x359.jpg&hash=8aad1b249c423654ae11d603959c574831452070)
In the first row the spectra for A, E, I, O, U are shown, person 1.
In the second row the spectra for A, E, I, O, U are shown, person 2.
In the third row the spectra for A, E, I, O, U are shown, person 3.
Although the fundamental frequencies of the vowels of the first person (in the first row) are much smaller than those of the two children (in the second and third rows), the spectral patterns for the individual vowels are reasonably similar. This is due to the resonance characteristics of the vocal tract, where each vowel can be identified in specific frequency bands (the formants) due to its spectral distribution.