Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition
Reexamination Certificate
2005-05-17
2005-05-17
McFadden, Susan (Department: 2655)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Recognition
C704S255000, C704S203000
Reexamination Certificate
active
06895376
ABSTRACT:
A reduced dimensionality eigenvoice analytical technique is used during training to develop context-dependent acoustic models for allophones. Re-estimation processes are performed to more strongly separate speaker-dependent and speaker-independent components of the speech model. The eigenvoice technique is also used during run time upon the speech of a new speaker. The technique removes individual speaker idiosyncrasies, to produce more universally applicable and robust allophone models. In one embodiment the eigenvoice technique is used to identify the centroid of each speaker, which may then be “subtracted out” of the recognition equation.
REFERENCES:
patent: 5046099 (1991-09-01), Nishimura
patent: 5664059 (1997-09-01), Zhao
patent: 5864810 (1999-01-01), Digalakis et al.
patent: 6073096 (2000-06-01), Gao et al.
patent: 6141644 (2000-10-01), Kuhn et al.
patent: 6327565 (2001-12-01), Kuhn et al.
patent: 6343267 (2002-01-01), Kuhn et al.
patent: 6571208 (2003-05-01), Kuhn et al.
A. Acero et al., “Speaker and gender normalization for continous-density Hidden Markov Models,” IEEE ICASSP '96, vol. 1, pp. 342-345, May 1996.*
R. Kuhn et al., “Eigenvoices for speaker adaptation,” Int. Conf. Speech Language Processing '98, vol. 5, Nov. 30-Dec. 4, 1998, pp. 1771-1774.*
Kuhn et al., “Rapid speaker adaptation in eigenvoice space,” IEEE Transactions on Speech and Audio Processing, vol. 8, No. 6, pp. 695-707, Nov. 2000.*
Padmanabhan et al., “Speaker clustering and transformations for speaker adaptation in speech recognition systems,” IEEE Transactions on Speech and Audio Processing, vol. 6, No. 1, pp. 71-77, Jan. 1998.*
Hazen et al., “A Comparision of novel techniques for instantaneous speaker adaptation,” Proc. of Eurospeech97, pp. 2047-2050.
Junqua Jean-Claude
Kuhn Roland
Nguyen Patrick
Perronnin Florent
Harness Dickey & Pierce PLC
Matsushita Electric - Industrial Co., Ltd.
McFadden Susan
LandOfFree
Eigenvoice re-estimation technique of acoustic models for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Eigenvoice re-estimation technique of acoustic models for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Eigenvoice re-estimation technique of acoustic models for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3460794