SPE-L1: Discriminative Training for ASR |
| Session Type: Lecture |
| Time: Tuesday, April 21, 11:00 - 13:00 |
| Location: Plenary Hall, TICC |
| Session Chair: Erik McDermott, NTT Corporation |
| SPE-L1.1: MODIFIED MPE/MMI IN A TRANSDUCER-BASED FRAMEWORK |
| Georg Heigold; RWTH Aachen University |
| Ralf Schlueter; RWTH Aachen University |
| Hermann Ney; RWTH Aachen University |
| SPE-L1.2: LARGE MARGIN SEMI-TIED COVARIANCE TRANSFORMS FOR DISCRIMINATIVE TRAINING |
| George Saon; IBM T. J. Watson Research Center |
| Daniel Povey; IBM T. J. Watson Research Center |
| Hagen Soltau; IBM T. J. Watson Research Center |
| SPE-L1.3: A TRUST REGION BASED OPTIMIZATION FOR MAXIMUM MUTUAL INFORMATION ESTIMATION OF HMMS IN SPEECH RECOGNITION |
| Zhi-Jie Yan; iFlytek Speech Lab, University of Science and Technology of China |
| Cong Liu; iFlytek Speech Lab, University of Science and Technology of China |
| Yu Hu; iFlytek Speech Lab, University of Science and Technology of China |
| Hui Jiang; York University |
| SPE-L1.4: LATTICE-BASED OPTIMIZATION OF SEQUENCE CLASSIFICATION CRITERIA FOR NEURAL-NETWORK ACOUSTIC MODELING |
| Brian Kingsbury; IBM |
| SPE-L1.5: BAYESIAN LARGE MARGIN HIDDEN MARKOV MODELS FOR SPEECH RECOGNITION |
| Jung-Chun Chen; National Cheng Kung University |
| Jen-Tzung Chien; National Cheng Kung University |
| SPE-L1.6: GENERALIZED BAUM-WELCH ALGORITHM FOR DISCRIMINATIVE TRAINING ON LARGE VOCABULARY CONTINUOUS SPEECH RECOGNITION SYSTEM |
| Roger Hsiao; InterACT, Language Technologies Institute, Carnegie Mellon University |
| Yik-Cheung Tam; InterACT, Language Technologies Institute, Carnegie Mellon University |
| Tanja Schultz; InterACT, Language Technologies Institute, Carnegie Mellon University |