Title :
On the effects of short-term spectrum smoothing in channel normalization
Author :
Avendano, Carlos ; Hermansky, Hynek
Author_Institution :
Dept. of Electr. Eng., Oregon Graduate Inst. of Sci. & Technol., Portland, OR, USA
fDate :
7/1/1997 12:00:00 AM
Abstract :
We present a simple analysis showing that channel normalization techniques are less effective when applied to spectral energies obtained by (weighted) summation of components of the short-time Fourier power spectrum of speech. We show that applying channel normalization processing prior to critical band integration or linear predictive all-pole modeling improves the effectiveness of the techniques
Keywords :
Fourier transforms; linear predictive coding; smoothing methods; spectral analysis; speech coding; speech processing; speech recognition; telecommunication channels; LPC features; automatic speech recognition; channel normalization; channel normalization processing; critical band integration; linear predictive all-pole modeling; short-term spectrum smoothing; short-time Fourier power spectrum; spectral energies; speech spectrum; weighted summation; Automatic speech recognition; Band pass filters; Cepstral analysis; Collision mitigation; Computational modeling; Frequency; Predictive models; Shape; Smoothing methods; Speech analysis;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
