Method and apparatus for speech recognition using optimized partial mixture tying of HMM state functions
US5825978A · kind A · utility
Assignee
Inventors
Key dates
| Filing date | Jul 18, 1994 |
| Grant date | Oct 20, 1998 |
| Priority date | — |
| Expiry date | Jul 18, 2014 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L15/146
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
In accordance with the invention, a speech recognizer is provided which uses a computationally-feasible method for constructing a set of Hidden Markov Models (HMMs) for speech recognition that utilize a partial and optimal degree of mixture tying. With partially-tied HMMs, improved recognition accuracy of a large vocabulary word corpus as compared to systems that use fully-tied HMMs is achieved with less computational overhead than with a fully untied system. The computationally-feasible technique comprises the steps of determining a cluster of HMM states that share Gaussian components which are close together, developing a subset codebook for those clusters, and recalculating the Gaussians in the codebook to best estimate the clustered states.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.