Method of active learning for automatic speech recognition

Details Method of active learning for automatic speech recognition Method of active learning for automatic speech recognition

2006-12-12

2006-12-12

Knepper, David D. (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

Reexamination Certificate

active

07149687

ABSTRACT:
State-of-the-art speech recognition systems are trained using transcribed utterances, preparation of which is labor-intensive and time-consuming. The present invention is an iterative method for reducing the transcription effort for training in automatic speech recognition (ASR). Active learning aims at reducing the number of training examples to be labeled by automatically processing the unlabeled examples and then selecting the most informative ones with respect to a given cost function for a human to label. The method comprises automatically estimating a confidence score for each word of the utterance and exploiting the lattice output of a speech recognizer, which was trained on a small set of transcribed data. An utterance confidence score is computed based on these word confidence scores; then the utterances are selectively sampled to be transcribed using the utterance confidence scores.

REFERENCES:
patent: 5513298 (1996-04-01), Stanford et al.
patent: 5606644 (1997-02-01), Chou et al.
patent: 5689616 (1997-11-01), Li
patent: 5839106 (1998-11-01), Bellegarda
patent: 5995930 (1999-11-01), Hab-Umbach et al.
patent: 6009392 (1999-12-01), Kanevsky et al.
patent: 6185530 (2001-02-01), Ittycheriah et al.
patent: 6377921 (2002-04-01), Bahl et al.
patent: 6385579 (2002-05-01), Padmanabhan et al.
patent: 6477493 (2002-11-01), Brooks et al.
patent: 6609093 (2003-08-01), Gopinath et al.
Teresa M. Kamm and Gerard G.L. Meyer, “Selective Sampling of Training Data For Speech”, in Proc. Human Language Technology, San Diego, CA 2002.
D. Cohn, L. Atlas, and R. Ladner, “Improving Generalization with Active Learning,” Machine Learning, vol. 15 pp. 201-112, 1994.
D.D. Lewis and J. Catlett, “Heterogenous Uncertainty Sampling for Supervised Learning,” in Proc. of the 11thInernational Conference on Machine Learning, 1994, pp. 148-156.
I. Dagan and S. P. Engelson, Committee-based Sampling for Training Probabilistic Classifiers, in Proc. of the 12thInternational Conference on Machine Learning, 1995, pp. 150-157.
C. Thompson, M. E. Califf, and R.J. Mooney, “Active Learning For Natural Language Parsing And Information Extraction,” in Proc. of the 16thInternational Conference on Machine Learning, 1999, pp. 406-414.
Y. Yang, “Sampling Strategies and Learning Efficiency in Text Categorization,” in Proc. of the AAAI Spring Symposium on Machine Learning in Information Access, M. Hearst and H. Hirst, Eds. 1996, pp. 88-95.
R.C. Rose, B.H. Juang, and C. H. Lee, “A Training Procedure for Verifying String Hypotheses in Continuous Speech Recognition,” in Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing, 1995, pp. 281-284.
R. Zhang and A. Rudnicky, “Word Level Confidence Annotation Using Combinations of Features,” in Proc. of 7thEuropean conference on speech communication and Technology, 2001, pp. 2105-2108.
Roberto Gretter and Giuseppe Riccardi, “On-Line Learning of Language Models with Word Error Probability Distributions,” in Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2001.
L. Mangu, E. Brill, and A. Stolcke, “Finding Consensus in Speech Recognition: Word Error Minimization and Other Applications of Confusion Networks,” Computer Speech and Language, vol. 14, No. 4, pp. 373-400, 2000.
A. Gorin, J.H. Wright, G. Riccardi, A. Abella, and T. Alonso, “Semantic Information Processing of Spoken Language,” in Proc. of ATR Workshop on Multi-Lingual Speech Communication, 2000.

Affiliated with

Also associated with

Rating

Method of active learning for automatic speech recognition does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method of active learning for automatic speech recognition, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of active learning for automatic speech recognition will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3717645