An analysis of cepstral-time feature matrices for noise and channel robust speech recognition

Ben Milner; Saeed V. Vaseghi

doi:10.21437/Eurospeech.1995-138

An analysis of cepstral-time feature matrices for noise and channel robust speech recognition

Ben Milner, Saeed V. Vaseghi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

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.,

Original language	English
Title of host publication	Proc. 4th European Conference on Speech Communication and Technology
DOIs	https://doi.org/10.21437/Eurospeech.1995-138
Publication status	Published - 1995
Event	Eurospeech 1995 - Duration: 1 Sep 1995 → …

Conference

Conference	Eurospeech 1995
Period	1/09/95 → …

Access to Document

10.21437/Eurospeech.1995-138

Cite this

@inproceedings{68afd2675cf54323b2abc40702c01fc6,

title = "An analysis of cepstral-time feature matrices for noise and channel robust speech recognition",

abstract = "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.,",

author = "Ben Milner and Vaseghi, {Saeed V.}",

year = "1995",

doi = "10.21437/Eurospeech.1995-138",

language = "English",

booktitle = "Proc. 4th European Conference on Speech Communication and Technology",

note = "Eurospeech 1995 ; Conference date: 01-09-1995",

}

TY - GEN

T1 - An analysis of cepstral-time feature matrices for noise and channel robust speech recognition

AU - Milner, Ben

AU - Vaseghi, Saeed V.

PY - 1995

Y1 - 1995

N2 - 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.,

AB - 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.,

U2 - 10.21437/Eurospeech.1995-138

DO - 10.21437/Eurospeech.1995-138

M3 - Conference contribution

BT - Proc. 4th European Conference on Speech Communication and Technology

T2 - Eurospeech 1995

Y2 - 1 September 1995

ER -