Active learning for spoken language understanding

Details Active learning for spoken language understanding Active learning for spoken language understanding

2007-08-28

2007-08-28

Hudspeth, David (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

C245S004000

Reexamination Certificate

active

10404699

ABSTRACT:
Disclosed is a system and method of training a spoken language understanding module. Such a module may be utilized in a spoken dialog system. The method of training a spoken language understanding module comprises training acoustic and language models using a small set of transcribed data ST, recognizing utterances in a set Suthat are candidates for transcription using the acoustic and language models, computing confidence scores of the utterances, selecting k utterances that have the smallest confidence scores from Suand transcribing them into a new set Si, redefining Stas the union of Stand Si, redefining Suas Suminus Si, and returning to the step of training acoustic and language models if word accuracy has not converged.

REFERENCES:
patent: 5860063 (1999-01-01), Gorin et al.
patent: 6173261 (2001-01-01), Arai et al.
patent: 2004/0205482 (2004-10-01), Basu et al.
Thompson et al. “Active Learning for Natural Language Parsing and Information Extraction”, Proc. Of 16th Intern. Machine Learning Conf., pp. 406-414, Jun. 1999.
C. Allauzen and M. Mohri. Efficient Algorithms for Testing the Twins Property.Journal of Automata, Languages and Combinatorics,2003.
M. Mohri. On Some Applications of Finite-State Automata Theory to Natural Language Processing.Journal of Natural Language Engineering,2:1-20, 1996.
M. Mohri. Finite-State Transducers in Language and Speech Processing. Computational Linguistics, 23(2), 1997.
M. Mohri, F.C.N. Pereira, and M. Riley. Weighted Automata in Text and Speech Processing. InProceedings of the 12thbiennial European Conference on Artificial Intelligence(ECAI-96),Workshop on Extended finite state models of language, Budapest, Hungary. ECAI, 1996.
M. Mohri and M. Riley. Integrated Context-Dependent Networks in Very Large Vocabulary Speech Recognition. InProceedings of the 6thEuropean Conference on Speech Communication and Technology(Eurospeech '99), Budapest, Hungary, 1999.

Affiliated with

Also associated with

Rating

Active learning for spoken language understanding does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Active learning for spoken language understanding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Active learning for spoken language understanding will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3893550