|
Authors: | Tobias Kaufmann, Thomas Ewender, Beat Pfister |
Group: | Computer Engineering |
Type: | Inproceedings |
Title: | Improving broadcast news transcription with a precision grammar and discriminative reranking |
Year: | 2009 |
Month: | September |
Pub-Key: | KEP09 |
Book Titel: | Proceedings of Interspeech |
Pages: | 356-359 |
Keywords: | speech processing |
Abstract: | We propose a new approach of integrating a precision grammar into speech recognition. The approach is based on a novel robust parsing technique and discriminative reranking. By reranking 100-best output of the LIMSI German broadcast news transcription system we achieved a significant reduction of the word error rate by 9.6% relative. To our knowledge, this is the first significant improvement for a real-world road-domain speech recognition task due to a precision grammar. |
Location: | Brighton (UK) |
Resources: | [BibTeX] [Paper as PDF] |