Source normalization training for HMM modeling of speech

Details Source normalization training for HMM modeling of speech Source normalization training for HMM modeling of speech

Filed on

1998-08-15

Date issued

2000-11-21

Examiner

Isen, Forester W.

Patent Class

Data processing: speech signal processing, linguistics, language

SubClass

Speech signal processing

SubSubClass

Recognition

Other Related Categories

704240, 704234, G10L 1514

Type

Patent

Status

active

Patent number

061515736

Description

ABSTRACT:
A maximum likelihood (ML) linear regression (LR) solution to environment normalization is provided where the environment is modeled as a hidden (non-observable) variable. By application of an expectation maximization algorithm and extension of Baum-Welch forward and backward variables (Steps 23a-23d) a source normalization is achieved such that it is not necessary to label a database in terms of environment such as speaker identity, channel, microphone and noise type.

REFERENCES:
patent: 5222146 (1993-06-01), Bahal et al.
patent: 5727124 (1998-03-01), Lee et al.
Anastasakos et al., A Compact Model for Speaker-Adaptive Training, BBN Systems and Technologies, pp. 1137-1140, Oct. 1996.
Jun Ishii and Masahiro Tonomura, "Speaker Normalization and Adaptation Based on Linear Transformation," Proceedings IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 2, 21-24, pp. 1055-1058, Apr. 1997.
Tasos Anastasakos et al., "A Compact Model for Speaker-Adaptive Training," Proceedings International Conference on Spoken Language Processing, vol. 2, 3-6, pp. 1137-1140, Oct. 1996.
Alejandro Acero et al., "Speaker and Gender Normalization for Continuous-Density Hidden Markov Models," Proceedings IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 1, 7-10, pp. 342-345, May 1996.

Affiliated with

Also associated with

Rating

Source normalization training for HMM modeling of speech does not yet have a rating. At this time, there are no reviews or comments for this patent.

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

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1266494