Speech recognition method including biased principal components

Electrical audio signal processing systems and devices – One-way audio signal program distribution – Public address system

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

3645135, 381 42, G10L 500

Patent

active

047180935

ABSTRACT:
A speech recognition method and apparatus employ a speech processing circuitry for repetitively deriving from a speech input, at a frame repetition rate, a plurality of acoustic parameters. The acoustic parameters represent the speech input signal for a frame time. A plurality of template matching and cost processing circuitries are connected to a system bus, along with the speech processing circuitry, for determining, or identifying, the speech units in the input speech, by comparing the acoustic parameters with stored template patterns. The apparatus can be expanded by adding more template matching and cost processing circuitry to the bus thereby increasing the speech recognition capacity of the apparatus. The speech processing circuitry establishes overlapping time durations for generating the acoustic parameters and further employs a sinc-Kaiser smoothing function in combination with a folding technique for providing a discrete Fourier transform. The Fourier spectra are transformed using a biased principal component analysis which optimizes the across class variance. The template matching and cost processing circuitries provide distributed processing, on demand, of the acoustic parameters for generating through a dynamic programming technique the recognition decision.

REFERENCES:
patent: 4032711 (1977-06-01), Sambur
patent: 4038503 (1977-07-01), Moshier
patent: 4100370 (1978-07-01), Suzuki et al.
patent: 4156868 (1979-05-01), Levinson
patent: 4227177 (1980-10-01), Moshier
patent: 4277644 (1981-07-01), Levinson et al.
patent: 4336421 (1982-06-01), Welch et al.
patent: 4426551 (1984-01-01), Komatsu et al.
patent: 4446531 (1984-05-01), Tanaka
patent: 4481593 (1984-11-01), Bahlor
patent: 4587670 (1986-05-01), Levinson et al.
patent: 4592085 (1986-05-01), Watari et al.
ICASSP 81, Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 3, Mar. 30-Apr. 1, 1981, Atlanta, Ga., pp. 1176-1179, IEEE, New York, U.S.; E. C. Bronson et al, "A Parallel Architecture for Speech Understanding".
ICASSP 82, Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 3, May 3-5, 1982, Paris, France, pp. 1629-1632, IEEE, New York, U.S.; P. F. Brown et al, "Partial Traceback and Dynamic Programming".
ICASSP 82, Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 2, May 3-5, 1982, Paris, France, pp. 903-906, IEEE, New York, U.S.; C. Vicenzi et al, "Multiprocessor Architecture for Real-Time Speech Recognition Systems".
Comptes Rendus Hebdomadaires Des Seance De L'Academie Des Sciences, vol. 273, No. 10, Serie A, Sep. 6, 1971, pp. 415-418, Paris, France; J. P. Haton et al, "Informatique Extraction de Parametres et Compression de L'Information Application a la Reconnaissance de la Parole".
IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-23, No. 1, Feb. 1975, pp. 67-72, IEEE, New York, U.S.; F. Itakura, "Minimum Prediction Residual Principle Applied to Speech Recognition".
5th International Conference on Pattern Recongition, vol. 1, Dec. 1-4, 1980, Florida, pp. 321-325, IEEE, New York; D. Dours et al, "Multiprocessor Architecture Adapted to the Parallel Treatment of a Continuous Flow of Data".

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Speech recognition method including biased principal components 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 method including biased principal components, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech recognition method including biased principal components will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-12667

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.