Optimization of perceptually-based ASR front-end (automatic speech recognition)

Several recently proposed automatic speech recognition (ASR) front-ends are experimentally compared for speaker-dependent and cross-speaker ASR. The perceptually based linear predictive front-end yields the highest accuracies. By modifying its sensitivity to spectral peaks and to spectral tilt and b...

Full description

Saved in:
Bibliographic Details
Published inICASSP-88., International Conference on Acoustics, Speech, and Signal Processing pp. 219 - 222 vol.1
Main Authors Hermansky, H., Junqua, J.C.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1988
Subjects
Automatic speech recognition
Bandwidth
Cepstral analysis
Filters
Frequency
Laboratories
Predictive models
Psychology
Speech analysis
Online AccessGet full text

Cover

More Information
Summary:Several recently proposed automatic speech recognition (ASR) front-ends are experimentally compared for speaker-dependent and cross-speaker ASR. The perceptually based linear predictive front-end yields the highest accuracies. By modifying its sensitivity to spectral peaks and to spectral tilt and by utilizing the speech dynamics the authors further improve, by about 10%, its error rate in speaker-independent ASR.< >
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.1988.196553