Improving broadcast news transcription with a precision grammar and discriminative reranking

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 broad-domain speech recognition task due to a precision grammar.

Extracted Key Phrases

3 Figures and Tables

Cite this paper

@inproceedings{Kaufmann2009ImprovingBN, title={Improving broadcast news transcription with a precision grammar and discriminative reranking}, author={Tobias Kaufmann and Thomas Ewender and Beat Pfister}, booktitle={INTERSPEECH}, year={2009} }