Characterizing Feature Variability in Automatic Speech Recognition Systems

Author

Barrault, Loic ; Matrouf, Driss ; de Mori, Renato ; Gemello, Roberto ; Mana, Franco

Author_Institution

LIA, Avignon

Volume

5

fYear

2006

fDate

14-19 May 2006

Abstract

A method is described for predicting acoustic feature variability by analyzing the consensus and relative entropy of phoneme posterior probability distributions obtained with different acoustic models having the same type of observations. Variability prediction is used for diagnosis of automatic speech recognition (ASR) systems. When errors are likely to occur, different feature sets are considered for correcting recognition results. Experimental results are provided on the CH1 Italian portion of AURORA3

Keywords

speech recognition; statistical distributions; AURORA3; CH1 Italian portion; acoustic feature variability prediction; acoustic models; automatic speech recognition systems; consensus analysis; feature variability characterization; phoneme posterior probability distributions; relative entropy analysis; Acoustic measurements; Automatic speech recognition; Context modeling; Entropy; Error analysis; Error correction; Hidden Markov models; Interpolation; Predictive models; Probability distribution;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on

Conference_Location

Toulouse

ISSN

1520-6149

Print_ISBN

1-4244-0469-X

Type

conf

DOI

10.1109/ICASSP.2006.1661454

Filename

1661454