Speech recognition system using modifiable recognition threshold

Details Speech recognition system using modifiable recognition threshold Speech recognition system using modifiable recognition threshold

1997-11-24

1999-10-19

Hudspeth, David R.

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

704238, G10L 506

Patent

active

059704500

ABSTRACT:
A speech recognition system, in which partial reference patterns, and cumulative similarities of these patterns, are stored in a temporary pattern memory. The partial reference patterns are to be used as subjects of a similarity computation with an input speech pattern that has its feature quantities extracted by a speech analyzing unit. A counting unit counts partial reference patterns having corresponding cumulative similarities that are higher than a threshold value stored in a threshold memory. A threshold computing unit computes a threshold of pruning from a correspondence relation between the number of partial reference patterns that have corresponding cumulative similarities that exceed the threshold, and the threshold. A similarity computing unit computes a similarity, with respect to the feature quantities, of partial reference patterns with corresponding cumulative similarities that are greater than the threshold of pruning. A determining unit determines, as a result of recognition, a word which corresponds to one of the partial reference patterns with a highest similarity.

REFERENCES:
patent: 4624008 (1986-11-01), Vensko et al.
patent: 4783803 (1988-11-01), Baker et al.
patent: 4811399 (1989-03-01), Landell et al.
patent: 4977598 (1990-12-01), Doddington et al.
patent: 4980918 (1990-12-01), Bahl et al.
patent: 5140668 (1992-08-01), Hattori
patent: 5241619 (1993-08-01), Schwartz et al.
patent: 5349645 (1994-09-01), Zhao
patent: 5577162 (1996-11-01), Yamazaki
patent: 5651094 (1997-07-01), Takagi et al.
patent: 5712956 (1998-01-01), Yamada et al.
patent: 5749069 (1998-05-01), Komori et al.
patent: 5774848 (1998-06-01), Hattori
patent: 5806021 (1998-09-01), Chen et al.
Hiroaki Sakoe, et al, "A High Speed DP-Matching Algorithm Based on Frame Syncronization, Beam Search and Vector Quantization", pp. 1650-1659, 1988.
Jinichi Murakami, et al, "A Spontaneous Speech Recognition Algorithm with Pause and Filled Pause Procedure", pp. 71-78, 1994.
L.R. Rabiner, et al., "An Introduction to Hidden Markov Models", pp. 4-16, 1986.
Anonymous: Dynamic Thresholding Scheme For Fast Match IBM Technical Disclosure Bulletin, vol. 32, No. 9B, Feb. 1990, pp 236-237.
Khoda M: A Study of Modifying Pruning Strategies for DP Beam Search at a preset input frame Systems & Computers in Japan, vol. 24 No. 3, 1993, pp. 98-110, XP000504548.
Hiroaki Sakoe et al., "A High-Speed DP-Matching Algorithm based on fram synchronization, beam search and vector quantization", Systems and Computers in Japan, vol. 20, Nov. 1998, pp. 33-44, XP000128842.
S. Ortmanns et al, "Language-Model Look-Ahead for Large Vocabulary Speech Recognition", Proceedings of the ICSLP, pp. 1-4, 1996.
Volker Steinbiss et al, "Improvements in Beam Search", Proceedings of the ICSLP, pp. 2143-2146, 1994.

Affiliated with

Also associated with

Rating

Speech recognition system using modifiable recognition threshold does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Speech recognition system using modifiable recognition threshold, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech recognition system using modifiable recognition threshold will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2069230