Discriminative training for speaker and speech verification

Details Discriminative training for speaker and speech verification Discriminative training for speaker and speech verification

2005-12-20

2008-11-18

Hudspeth, David R (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

C704S239000, C704S246000

Reexamination Certificate

active

07454339

ABSTRACT:
A method for discriminatively training acoustic models is provided for automated speaker verification (SV) and speech (or utterance) verification (UV) systems. The method includes: defining a likelihood ratio for a given speech segment, whose speaker identity (for SV system) or linguist identity (for UV system) is known, using a corresponding acoustic model, and an alternative acoustic model which represents all other speakers (in SV) or all other linguist identities (in UV); determining an average likelihood ratio score for the likelihood ratio scores over a set of training utterances (referred to as true data set) whose speaker identities (for SV) or linguist identities (for UV) are the same; determining an average likelihood ratio score for the likelihood ratio scores over a competing set of training utterances which excludes the speech data in the true data set (referred to as competing data set); and optimizing a difference between the average likelihood ratio score over the true data set and the average likelihood ratio score over the competing data set, thereby improving the acoustic model.

REFERENCES:
patent: 5579436 (1996-11-01), Chou et al.
patent: 2003/0036904 (2003-02-01), Chaudhari et al.
Tamura, S. Iwano, K. Furai, S. “Improvement of audio-visual speech recognition in cars” in Proceedings of the 18th International Congress on Acoustics (ICA '04), vol. 4, pp. 2595-2598, Kyoto, Japan, Apr. 2004.
Wilcox, L. Chen, F. Kimber, D. Balasubramanian, V. “Segmentation of speech using speaker identification” Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on, vol. i, On pp. I/161-I/164.
H. Gish, M.-H. Siu, R. Rohlicek, “Segregation of speaker for speech recognition and speaker identification,” icassp, pp. 873-876, Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on, 1991.
J. Navratil, U.V. Chaudhari, G.N. Ramaswamy, “Speaker verification using target and background dependent linear transforms and multi-system fusion,” Proc. of Eurospeech-01, Aalborg, Denmark, Sep. 2001.
Sukkar et al, “Utterance Verification of Keyword Strings Using Word-Based Minimum Verification Error (WB-MVE) Training”, Proc. ICASSP'96, Atlanta May 1996.
Rosenberg et al, “Speaker Verification Using Minimum verificatio Error Training”, ICASSP'98, Seattle, May 1998.

Affiliated with

Also associated with

Rating

Discriminative training for speaker and speech verification does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Discriminative training for speaker and speech verification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Discriminative training for speaker and speech verification will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4025827