Automatic Speech Recognition with Confidence Measures Obtained by Multiple Recognizers of Various Recognition Units
| Project/Area Number |
21700184
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Doshisha University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2009 – 2011
|
| Project Status |
Completed (Fiscal Year 2011)
|
| Budget Amount *help |
¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2011: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2010: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2009: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
|
|---|
| Keywords | 音声認識 / 認識単位 / 複合語 / 信頼度 / Support Vector Machine / 音声対話 / SVM / 目的地名 |
|---|
| Research Abstract |
In this study, I proposed a novel speech recognition method that improves the word recognition accuracy by using confidence measures obtained from multiple recognizers of various recognition units. The method uses Support Vector Machine(SVM) to select a single recognition result from multiple recognition results. The method can also identify whether the correct recognition result is included in the multiple recognition results. Experimental results show that the proposed method gives higher word accuracy and classification accuracy in comparison with a single recognizer and ROVER.
|
Report
(4 results)
Research Products
(9 results)
