FLC-regulated speaker adaptation mechanisms for speech recognition

Authors:
Ing-Jr Ding
Affiliations:
Department of Electrical Engineering, National Formosa University, Huwei Township, Yunlin County, Taiwan, R.O.C.
Venue:
ICCCI'10 Proceedings of the Second international conference on Computational collective intelligence: technologies and applications - Volume Part II
Year:
2010

Citing 3
Cited 0

Speech recognition by machines and humans

Speech Communication
Telephone speech recognition based on Bayesian adaptation of hidden Markov models

Speech Communication
Kernel Eigenspace-Based MLLR Adaptation

IEEE Transactions on Audio, Speech, and Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The exploitation of fuzzy logic control (FLC) mechanism in the fields of speaker adaptation (SA) is thoroughly investigated in this study, specifically in the reliable determination of HMM acoustic parameters. For enhancing the performance of speaker adaptation, the FLC mechanism is engineered into the MAP estimate of HMM parameters for Bayesian-based adaptation; also into the MLLR estimate for transformation-based adaptation. The speech recognition system using an adaptation scheme with the support of FLC will still be able to keep a satisfactory recognition performance even in an ordinary case.