Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SP-L11.6 |
| Session: | Language Modeling and Search |
| Time: | Friday, May 21, 17:10 - 17:30 |
| Presentation: |
Lecture |
| Topic: |
Speech Processing: Language Modeling |
| Title: |
THE USE OF A LINGUISTICALLY MOTIVATED LANGUAGE MODEL IN CONVERSATIONAL SPEECH RECOGNITION |
| Authors: |
Wen Wang; SRI International / Purdue University | | |
| | Andreas Stolcke; SRI International | | |
| | Mary Harper; Purdue University | | |
| Abstract: |
Structured language models have recently been shown to give significant improvements in large-vocabulary recognition relative to standard N-gram models, but typically imply a heavy computational burden and have not been applied to large training sets or complex recognition systems. In previous work, we developed a linguistically motivated and computationally efficient almost-parsing language model using a data structure derived from Constraint Dependency Grammar parses that tightly integrates knowledge of words, lexical features, and syntactic constraints. In this paper we show that such a model can be used effectively and efficiently in all stages of a complex, multi-pass conversational telephone speech recognition system. Compared to a state-of-the-art 4-gram interpolated word- and class-based language model, we obtained a 6.2\% relative word error reduction (a 1.6\% absolute reduction) on a recent NIST evaluation set. |
| |
| Back | |