An efficient augmented-context-free parsing algorithm
Computational Linguistics
Construction of corpus-based syntactic rules for accurate speech recognition
COLING '92 Proceedings of the 14th conference on Computational linguistics - Volume 2
A neural fuzzy training approach for continuous speech recognition improvement
ICASSP'92 Proceedings of the 1992 IEEE international conference on Acoustics, speech and signal processing - Volume 1
A segment-based speaker adaptation neural network applied to continuous speech recognition
ICASSP'92 Proceedings of the 1992 IEEE international conference on Acoustics, speech and signal processing - Volume 1
A successive state splitting algorithm for efficient allophone modeling
ICASSP'92 Proceedings of the 1992 IEEE international conference on Acoustics, speech and signal processing - Volume 1
| Hi-index | 0.00 |
This paper describes ATREUS, an aggregation of a large variety of continuous speech recognition systems developed at ATR Interpreting Telephony Research Laboratories forming the spoken input front-end of an interpreting telephony system. ATREUS includes the following phone models: (1) discrete HMMs with fuzzy vector quantization and multiple codebooks, (2) continuous mixture density HMMs, (3) Hidden Markov networks derived from the Successive State Splitting algorithm, (4) Time-delay Neural Networks, and (5) Fuzzy Partition Models. Its speaker modes involve (a) speaker-dependent, (b) speaker- independent, and (c) speaker-adaptive techniques such as codebook mapping for VQ-HMMs, vector field smoothing for all types of HMMs, and neural network speaker mapping. ATREUS is one of the major achievements in the seven-year automatic interpreting telephony project, scheduled to end at the end of this fiscal year. A comparative study is given from the view points of structure, constituent techniques, hardware implementation and performance. A combination called ATREUS/SSS-LR had the best performance among the ATREUS systems.