Abstract
We describe our continuing efforts to improve the UW-SRI-ICSI Mandarin broadcast speech recognizer. This includes increasing acoustic and text training data, adding discriminative features, incorporating frame-level discriminative training criterion, multiple-pass acoustic model (AM) cross adaptation, language model (LM) genre adaptation and system combination. The net effect without LM adaptation was a 24%-64% relative reduction in character error rates (CERs) on a variety of test sets. In addition, LM adaptation gave us another 6% of relative CER reduction on broadcast conversations.
| Original language | English |
|---|---|
| Pages (from-to) | 2876-2879 |
| Number of pages | 4 |
| Journal | Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH |
| Volume | 2007 |
| DOIs | |
| Publication status | Published - 1 Dec 2007 |
| Externally published | Yes |
| Event | 8th Annual Conference of the International Speech Communication Association, Interspeech 2007 - Antwerp, Belgium Duration: 27 Aug 2007 → 31 Aug 2007 |
Keywords
- Character error rates
- Discriminative features
- Discriminative training
- LM adaptation
- Mandarin
ASJC Scopus subject areas
- Computer Science Applications
- Software
- Modelling and Simulation
- Linguistics and Language
- Communication
Fingerprint
Dive into the research topics of 'Advances in mandarin broadcast speech recognition'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver