This paper introduces recent research activities on speech recognition, ranging from acoustic processing to linguistic processing, at NTT (Nippon Telegraph and Telephone Corporation) Laboratories. These include the proposal of hierarchical cepstral parameters and ALSP parameters, a new method of utilizing pitch information, automatic speaker adaptation techniques, robust HMM phoneme models, new training algorithms for neural networks, linguistic processing using syntactic and semantic knowledge, implementation of prototype continuous speech recognition systems, and an efficient text-independent speaker recognition algorithm.
Cite as: Furui, S. (1991) Recent advances in speech recognition. Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991), 3-12, doi: 10.21437/Eurospeech.1991-1
@inproceedings{furui91_eurospeech,
author={Sadaoki Furui},
title={{Recent advances in speech recognition}},
year=1991,
booktitle={Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991)},
pages={3--12},
doi={10.21437/Eurospeech.1991-1}
}