Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

An Analysis of Cepstral-Time Matrices for Noise and Channel Robust Speech Recognition

Ben P. Milner, Saeed V. Vaseghi

School of Information Systems, University of East Anglia, Norwich, UK

This paper presents an analysis of the cepstral-time matrix. The coefficients of the cepstral-time matrix are found to be similar to the standard cepstral vector with differential features augmented on. It is also shown that the cepstral-time matrix is inherently robust to convolutional channel distortion. Spectral-subtraction, Wiener filtering and model combination are extended into two-dimensions where improved noise robustness is achieved. Experimental results using the NOISEX database with noise and channel distorted speech are presented.,

Full Paper

Bibliographic reference.  Milner, Ben P. / Vaseghi, Saeed V. (1995): "An analysis of cepstral-time matrices for noise and channel robust speech recognition", In EUROSPEECH-1995, 519-522.