Method and apparatus for training language model, and method and apparatus for recognizing language
US10380483B2 · kind B2 · utility
1Cited by
13References
21Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Jan 11, 2016 |
| Grant date | Aug 13, 2019 |
| Priority date | — |
| Expiry date | Mar 18, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L15/005
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A method and apparatus for training a language model, include generating a first training feature vector sequence and a second training feature vector sequence from training data. The method is configured to perform forward estimation of a neural network based on the first training feature vector sequence, and perform backward estimation of the neural network based on the second training feature vector sequence. The method is further configured to train a language model based on a result of the forward estimation and a result of the backward estimation.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.