Speech and speaker recognition using factor analysis to model co

Details Speech and speaker recognition using factor analysis to model co Speech and speaker recognition using factor analysis to model co

1997-11-17

1999-08-31

Knepper, David D.

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

704240, G01L 916

Patent

active

059466561

ABSTRACT:
Hidden Markov models (HMMs) rely on high-dimensional feature vectors to summarize the short-time properties of speech correlations between features that can arise when the speech signal is non-stationary or corrupted by noise. These correlations are modeled using factor analysis, a statistical method for dimensionality reduction. Factor analysis is used to model acoustic correlation in automatic speech recognition by introducing a small number of parameters to model the covariance structure of a speech signal. The parameters are estimated by an Expectation Maximization (EM) technique that can be embedded in the training procedures for the HMMs, and then further adjusted using Minimum Classification Error (MCE) training, which demonstrates better discrimination and produces more accurate recognition models.

REFERENCES:
patent: 4038503 (1977-07-01), Moshier
patent: 5473728 (1995-12-01), Luginbuhl et al.
patent: 5555344 (1996-09-01), Zunkler
patent: 5579436 (1996-11-01), Chou et al.
patent: 5590242 (1996-12-01), Juang et al.
Ghahramani et al. "The EM Algorithm for Mixtures of Factor Analyzers", Technical Report CRG-TR-96-1, May 21, 1996, pp. 1-8. University of Toronto.
Hinton et al. "Modelling the Manifolds of Images of Handwritten Digits", Dept. of Computer Science, (IEEE trans. on Neural Networks), pp. 1-29.
E. Gunter Schukat-Talamazzini, Joachim Hornegger, and Heinrich Niemann, "Optimal Linear Feature Transformations for Semi-Continuous Hidden Markov Models," Proc. IEEE ICASSP 95, vol. 1, pp. 369-372, May 1995.

Affiliated with

Also associated with

Rating

Speech and speaker recognition using factor analysis to model co does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Speech and speaker recognition using factor analysis to model co, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech and speaker recognition using factor analysis to model co will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2428823