Abstract | ||
---|---|---|
We describe the development of the ICSI-SRI speech recognition system for the National Institute of Standards and Technology (NIST) Spring 2006 Meeting Rich Transcription (RT-06S) evaluation, highlighting improvements made since last year, including improvements to the delay-and-sum algorithm, the nearfield segmenter, language models, posterior-based features, HMM adaptation methods, and adapting to a small amount of new lecture data. Results are reported on RT-05S and RT-06S meeting data. Compared to the RT-05S conference system, we achieved an overall improvement of 4% relative in the MDM and SDM conditions, and 11% relative in the IHM condition. On lecture data, we achieved an overall improvement of 8% relative in the SDM condition, 12% on MDM, 14% on ADM, and 15% on IHM. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1007/11965152_39 | MLMI |
Keywords | Field | DocType |
icsi-sri spring,national institute,hmm adaptation method,rt-05s conference system,new lecture data,rt-06s meeting data,icsi-sri speech recognition system,ihm condition,lecture data,overall improvement,meeting recognition system,sdm condition,speech recognition,near field,language model | Recognition system,Computer science,Word error rate,Speech recognition,Maximum likelihood linear regression,NIST,Artificial intelligence,Language model,Machine learning,Acoustic model | Conference |
Volume | ISSN | ISBN |
4299 | 0302-9743 | 3-540-69267-3 |
Citations | PageRank | References |
12 | 1.62 | 11 |
Authors | ||
7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Adam Janin | 1 | 250 | 34.11 |
Andreas Stolcke | 2 | 6690 | 712.46 |
Xavier Anguera | 3 | 624 | 54.28 |
Kofi Boakye | 4 | 155 | 13.64 |
Özgür Çetin | 5 | 154 | 14.41 |
Joe Frankel | 6 | 312 | 22.78 |
Jing Zheng | 7 | 442 | 43.00 |