Abstract | ||
---|---|---|
A speech interface is often required in many application envi- ronments, such as telephone-based information retrieval, car navigation systems, and user-friendly interfaces, but the low speech recognition rate makes it difficult to extend its appli- cation to new fields. We propose a domain adaptation tech- nique via error correction with a maximum entropy language model, which is a general and elegant framework to com- bine higher level linguistic knowledge. Our approach has the ability to correct both semantic and lexical errors in 1-best output from the black-box style speech recognizer, and can improve the performance of speech recognition and applica- tion system. Through extensive experiments using a speech- driven in-vehicle telematics information retrieval and spoken language understanding, we demonstrate the superior perfor- mance of our approach and some advantages over previous lexical-oriented error correction approaches. |
Year | Venue | Keywords |
---|---|---|
2004 | INTERSPEECH | information retrieval,language model,error correction,maximum entropy,speech recognition |
Field | DocType | Citations |
Cache language model,Computer science,Car navigation systems,Natural language processing,Artificial intelligence,Language model,Spoken language,Factored language model,Pattern recognition,Speech recognition,Error detection and correction,Principle of maximum entropy,Telematics | Conference | 13 |
PageRank | References | Authors |
0.97 | 6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sangkeun Jung | 1 | 197 | 15.23 |
Minwoo Jeong | 2 | 142 | 13.89 |
Gary Geunbae Lee | 3 | 932 | 93.23 |