Resonant modes of otoacoustic emissions.

Transiently evoked otoacoustic emission (TEOAE) signals were decomposed into basic components by means of the matching pursuit algorithm. The components were characterized by frequency, latency, time span and energy. The resonant modes characteristic for each person/ear were identified. A high time-frequency resolution of the method and the parametric representation of the waveforms made possible estimation of the frequency ratios of the resonant modes. The repetitive occurrence of the 'fifth', 'fourths' and octaves connected with the Pythagorean temperament was found. The significance of the results was confirmed by comparison to Monte Carlo simulations of the null hypothesis of a random distribution of frequency modes. These findings are in agreement with the resonance theory of hearing, which binds musical ratios with the geometrical spacing of outer hair cells in cochlea.