MMSE estimation of log-filterbank energies for robust speech recognition

There are no files associated with this record.

Title MMSE estimation of log-filterbank energies for robust speech recognition
Author Stark, Anthony Phillip; Paliwal, Kuldip Kumar
Journal Name Speech Communication
Year Published 2011
Place of publication Netherlands
Publisher Elsevier BV * North-Holland
Abstract In this paper, we derive a minimum mean square error log-filterbank energy estimator for environment-robust automatic speech recognition. While several such estimators exist within the literature, most involve trade-offs between simplifications of the log-filterbank noise distortion model and analytical tractability. To avoid this limitation, we extend a well known spectral domain noise distortion model for use in the log-filterbank energy domain. To do this, several mathematical transformations are developed to transform spectral domain models into filterbank and log-filterbank energy models. As a result, a new estimator is developed that allows for robust estimation of both log-filterbank energies and subsequent Mel-frequency cepstral coefficients. The proposed estimator is evaluated over the Aurora2, and RM speech recognition tasks, with results showing a significant reduction in word recognition error over both baseline results and several competing estimators.
Peer Reviewed Yes
Published Yes
Alternative URI
Volume 53
Issue Number 3
Page from 403
Page to 416
ISSN 0167-6393
Date Accessioned 2012-03-20; 2012-04-10T23:50:26Z
Research Centre Institute for Integrated and Intelligent Systems
Faculty Faculty of Science, Environment, Engineering and Technology
Subject Artificial Intelligence and Image Processing
Publication Type Journal Articles (Refereed Article)
Publication Type Code c1

Show simple item record

Griffith University copyright notice