Development of the 2008 SRI Mandarin Speech-To-Text System for Broadcast News and Conversation

Citation

Lei, X., Wu, W., Wang, W., Mandal, A., & Stolcke, A. Development of the 2008 SRI Mandarin Speech-to-text System for Broadcast.

Abstract

We describe the recent progress in SRI’s Mandarin speech-to-text system developed for 2008 evaluation in the DARPA GALE program. A data-driven lexicon expansion technique and language model adaptation methods contribute to the improvement in recognition performance. Our system yields 8.3 pct. character error rate on the GALE dev08 test set, and 7.5 pct. after combining with RWTH systems. Compared to our 2007 evaluation system, a significant improvement of 13 pct. relative has been achieved.


Read more from SRI