Nonlinear mapping for feature extraction in automatic speech recognition
US7254538B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Nov 16, 2000 |
| Grant date | Aug 7, 2007 |
| Priority date | — |
| Expiry date | Aug 16, 2024 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L15/144
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
The present invention successfully combines neural-net discriminative feature processing with Gaussian-mixture distribution modeling (GMM). By training one or more neural networks to generate subword probability posteriors, then using transformations of these estimates as the base features for a conventionally-trained Gaussian-mixture based system, substantial error rate reductions may be achieved. The present invention effectively has two acoustic models in tandem—first a neural net and then a GMM. By using a variety of combination schemes available for connectionist models, various systems based upon multiple features streams can be constructed with even greater error rate reductions.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.