| 发明名称 |
Methods and systems for obtaining language models for transcribing communications |
| 摘要 |
A method for producing speech recognition results on a device includes receiving first speech recognition results, obtaining a language model, wherein the language model represents information stored on the device, and using the first speech recognition results and the language model to generate second speech recognition results. |
| 申请公布号 |
US9099087(B2) |
申请公布日期 |
2015.08.04 |
| 申请号 |
US201113221897 |
申请日期 |
2011.08.30 |
| 申请人 |
Canyon IP Holdings, LLC |
发明人 |
Adams Jeffrey P.;Basye Kenneth;Thomas Ryan;O'Neill Jeffrey C. |
| 分类号 |
G10L15/18;G10L15/19 |
主分类号 |
G10L15/18 |
| 代理机构 |
|
代理人 |
|
| 主权项 |
1. A method for producing speech recognition results on a receiving device, comprising:
receiving first speech recognition results from a sending device; obtaining a first language model from a source other than an individual user of the receiving device, wherein the first language model represents information stored on the receiving device, and wherein the first language model is associated with the individual user of the receiving device; generating a second language model by modifying the first language model based on the first speech recognition results; and using the first speech recognition results and the second language model to generate second speech recognition results. |
| 地址 |
Wilmington DE US |
