Modification of sub-phoneme speech spectral models for lombard s

Details Modification of sub-phoneme speech spectral models for lombard s Modification of sub-phoneme speech spectral models for lombard s

1997-10-23

1999-01-26

Hudspeth, David R.

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

704233, 704244, 704255, G10L 906

Patent

active

058648097

ABSTRACT:
A speech recognition apparatus recognizes utterances of unnatural speech having a higher performance of recognition accuracy with a smaller amount of speech learning data. The speech recognition apparatus includes an acoustic-phonetic variability learning unit, a normal speech model memory, a spectrum smoother-modifier and a speech recognizer. An input speech signal is acoustically analyzed and transformed into a time-series feature vector. The acoustic-phonetic variability learning unit learns an acoustic-phonetic change of spectrum caused by unnatural speech and generates a plurality of acoustic-phonetic variability models. The normal speech model memory stores a normal speech model learned based on normal speech data. The spectrum smoother-modifier modifies the normal speech model based on a plurality of the acoustic-phonetic variability model and generates a plurality of spectrum-modified speech models. The speech recognizer recognizes the time-series feature vector based on the normal speech model and the spectrum-modified speech model.

REFERENCES:
patent: 5313555 (1994-05-01), Kamiya
patent: 5361324 (1994-11-01), Takizawa et al.
patent: 5579436 (1996-11-01), Chou et al.
Satoru Hayomizu, Kazuyo Tanaka, and Kozo Ohta, "A Large Vocabulary Word Recognition System Using Rule-Based Network Representation of Acoustic Characteristic Variations," Proc. ICASSP 88, Paper S5.8, pp. 211-214, Apr. 1988.
Yoshiaki Itoh, Jiro Kiyama, and Ryuichi Oka, "Sentence Spotting Applied to Partial Sentences and Unknown Words," Proc. ICASSP 94, vol. 1, pp. 369-372, Apr. 1994.
Sang-Mun Chi, et al. "Lombard Effect Compensation and Noise Suppression for Noisy Lombard Speech Recognition," Proc. International Conference on Spoken Language Processing (ICSLP 96), Oct. 1996.
ICSLP 94 "Isolated Word Recognition Using Models for Acoustic Phonetic Variability by Lombard Effect" Suzuki, Nakajima & Abe, Computer & Information Systems Lab., Mitsubishi Electric--Japan.
Junqua, et al., "Acoustic and Perceptual Studies of Lombard Speech: Application to Isolated Words Automatic Speech Recognition" ICASSP '90:Acoustics, Speech Signal Processing Conference.
Hansen, J. "Morphological Constrained Feaeture Enhancement with Adaptive Cepstral Compensation (MCD-ACC) for Speech Recognition in Noise and Lombard Effect" IEEE Transactions on Speech . . . vol. 2 No. 4.
Junqua, J. "The Influence of Psychoacoustic and Psycholinguistic factors on Listner Judgments of Intelligibility of Normal & Lombard Speech" ICASSP '91: Acoustics, Speech & Signal Processing Conf.
Hanson, B. et al "Robust Speaker-Independent Word Recognition Using Static, Dynamic Acceleration Features: Experiments with Lombard & Noisy Speech" ICASSP '90: Acoustics, Speech & Signal Processing Conference.

Affiliated with

Also associated with

Rating

Modification of sub-phoneme speech spectral models for lombard s does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Modification of sub-phoneme speech spectral models for lombard s, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Modification of sub-phoneme speech spectral models for lombard s will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1458359