Data-driven codebook adaptation in phonetically tied SCHMMs
- 19 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1 (15206149) , 477-479
- https://doi.org/10.1109/icassp.1995.479632
Abstract
This paper reports the results of our experiments aimed at the automatic optimization of the number of parameters in the semi-continuous phonetically tied HMM based speech recognition system that is part of the speech-to-speech translation system JANUS-2. We propose different algorithms devised to determine the optimal number of model parameters. In recognition experiments performed on a spontaneous human-to-human dialog database, we show that automatic optimization of the acoustic modeling parameter size with the proposed algorithm improves the recognition performance without increasing the required amount of computing power and memory.Keywords
This publication has 5 references indexed in Scilit:
- JANUS 93: towards spontaneous speech translationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Learning state-dependent stream weights for multi-codebook HMM speech recognition systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- On semi-continuous hidden Markov modelingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Discriminant analysis and supervised vector quantization for continuous speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Acoustic and language modeling of human and nonhuman noises for human-to-human spontaneous speech recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002