9th Annual Conference of the International Speech Communication Association

Brisbane, Australia
September 22-26, 2008

The CMU-InterACT 2008 Mandarin Transcription System

Roger Hsiao, Mark Fuhs, Yik-Cheung Tam, Qin Jin, Tanja Schultz

Carnegie Mellon University, USA

We present our Mandarin BN/BC transcription system recently developed for the GALE07 evaluation. The system employs a 3-pass decoding strategy trained with over 1300 hours of quickly transcribed audio. We successfully apply discriminative training, dynamic unsupervised language model adaptation, and system combination techniques in our system. We furthermore achieve improvements by combining an Initial-Final system with a genre dependent phone system. On the GALE07 phase 2 retest evaluation, our system achieves a character error rate(CER) of 13.3% on dev07 test set and 13.5% on eval07 unsequestered test set. Our system also allows combination with other sites and in this paper, we investigate different system combination strategies which significantly improve the final recognition performance.

Full Paper

Bibliographic reference.  Hsiao, Roger / Fuhs, Mark / Tam, Yik-Cheung / Jin, Qin / Schultz, Tanja (2008): "The CMU-interACT 2008 Mandarin transcription system", In INTERSPEECH-2008, 1445-1448.