Cross-Validation EM Training for Robust Parameter Estimation
- 1 April 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4 (15206149) , IV-437-437
Abstract
A new maximum likelihood training algorithm is proposed that compensates for weaknesses of the EM algorithm by using cross-validation likelihood in the expectation step to avoid overtraining. By using a set of sufficient statistics associated with a partitioning of the training data, as in parallel EM, the algorithm has the same order of computational requirements as the original EM algorithm. Analyses using a GMM with artificial data show the proposed algorithm is more robust for overtraining than the conventional EM algorithm. Large vocabulary recognition experiments on Mandarin broadcast news data show that the method makes better use of more parameters and gives lower recognition error rates than EM training.Keywords
This publication has 7 references indexed in Scilit:
- Model Complexity Selection and Cross-Validation EM Training for Robust Speaker Diarization2007
- The IBM 2004 Conversational Telephony System for Rich Transcription2006
- Hmm State Clustering Based on Efficient Cross-Validation2006
- Maximum mutual information estimation of hidden Markov model parameters for speech recognition2005
- The Elements of Statistical Learning2001
- Acoustic modeling based on the MDL principle for speech recognition1997
- Flexible speaker adaptation for large vocabulary speech recognition1995