Investigations of issues for using multiple acoustic models to improve continuous speech recognition

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

Volumn 2, Issue , 2006, Pages 529-532

  Zhang, Rong ^a   Rudnicky, Alexander I ^a  

^a Carnegie Mellon University ^*  (United States)

Author keywords

Boosting; ROVER

Indexed keywords

ADAPTIVE BOOSTING; CONTINUOUS SPEECH RECOGNITION; SPEECH PROCESSING;

ACOUSTIC MODEL TRAININGS; ADABOOST ALGORITHM; BOOSTING; ENSEMBLE OF CLASSIFIERS; MEETING RECOGNITION; OBJECTIVE FUNCTIONS; ROVER; VOTING TECHNIQUES;

SPEECH RECOGNITION;

EID: 44949221930     PISSN: None     EISSN: 19909772     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (9)

1. th International Joint Conference on Artificial Intelligence, 1999.
2. J. Friedman, T. Hastie and R. Tibshirani, "Additive Logistic Regression: A Statistical View of Boosting", Technical Report, Stanford University, 1998.
3. L. Mason, J. Baxter, P. Bartlett, and M. Frean, "Boosting Algorithms as Gradient Descent", In Advances in Neural Information Processing Systems 12, pp 512-518, MIT Press, 2000.
4. R. Schluter, "Investigations on Discriminative Training Criteria", Ph.D. Thesis, Aachen, Germany, 2000.
5. I. Zitouni, H. K. J. Kuo, and C. H. Lee, "Combination of Boosting and Discriminative Training Techniques for Natural Language Call Steering Systems", Proc. of ICASSP, 2002.
6. J. G. Fiscus, "A Post-Processing System to Yield Reduced Word Error Rates: Recognizer Output Voting Error Reduction (ROVER)", Proc. of ASRU 1997.
7. F. Wessel, K. Macherey and R. Schluter, "Using Word Probabilities as Confidence Measures", Proc. of ICASSP 1998.
8. R. Zhang and A. I. Rudnicky, "A Frame Level Boosting Training Scheme for Acoustic Modeling", Proc. of ICSLP 2004.
9. S. Banerjee, J. Cohen, T. Quisel, et al, "Creating Multi-Modal, User-Centric Records of Meetings with the Carnegie Mellon Meeting Recorder Architecture", Proc. of ICASSP 2004 Meeting Recognition Workshop.