Re: Can a timbre be affected by a shifted (virtual) pitch evoked (Eckard Blumschein )

Subject: Re: Can a timbre be affected by a shifted (virtual) pitch evoked
From:    Eckard Blumschein  <Eckard.Blumschein(at)E-TECHNIK.UNI-MAGDEBURG.DE>
Date:    Tue, 25 Feb 2003 13:08:43 +0100

Thank you Chen-Gia Tsai, having looked at your wav file in MATLAB, I have to essentially correct myself: I still imagine the main process of hearing to proceed in two steps. It starts with spectral analysis within the organ of Corti. We may model it by FCT or abs(FFT). The next step is to once again apply IFCT or IFFT or FCT or FFT, however this time to the spectrum and without abs. In each case, the result is equivalent to the (causal) auto-convolution, in other words, to the auto-correlation of the sound signal itself. Application of auto-correlation to the spectrum would not make much sense. The neurons have to perform a second spectral analysis by selective inhibition or excitation, both based on the temporal structure of the already spectrally organized components of the signal. Eckard

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University