Title :
A novel speaker adaptation approach for continuous densities HMM´s
Author :
Frangoulis, E. ; Sgardoni, V.
Author_Institution :
Logica Cambridge Ltd., UK
Abstract :
An approach for speaker adaptation aiming to get high recognition performance from an HMM speech recognizer after a short training session with a new speaker is presented. The technique presented exploits the Gaussian multivariate nature of continuous density HMM distributions, to adapt the model parameters. This adaptation technique was applied to a 20-word vocabulary. It was tested on 70 new speakers after a training session of 2 to 5 repetitions of each vocabulary word. The experiments carried out have shown a significant improvement in the recognition performance, even when only two training tokens from the new speaker are used
Keywords :
Markov processes; speech recognition; HMM speech recognizer; continuous Gaussian mixtures; continuous densities HMM; continuous density HMM distributions; model parameters adaptation; recognition performance; speaker adaptation; training tokens; vocabulary; Bayesian methods; Databases; Hidden Markov models; Natural languages; Probability; Speech enhancement; Speech recognition; Testing; Training data; Vocabulary;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-0003-3
DOI :
10.1109/ICASSP.1991.150474