Method for speech processing involving whole-utterance modeling
US6961703B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 13, 2000 |
| Grant date | Nov 1, 2005 |
| Priority date | — |
| Expiry date | Oct 20, 2022 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L19/022
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A speech verification process involves comparison of enrollment and test speech data and an improved method of comparing the data is disclosed, wherein segmented frames of speech are analyzed jointly, rather than independently. The enrollment and test speech are both subjected to a feature extraction process to derive fixed-length feature vectors, and the feature vectors are compared, using a linear discriminant analysis and having no dependence upon the order of the words spoken or the speaking rate. The discriminant analysis is made possible, despite a relatively high dimensionality of the feature vectors, by a mathematical procedure provided for finding an eigenvector to simultaneously diagonalize the between-speaker and between-channel covariances of the enrollment and test data.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.