论文信息 - A Combined Task Analysis Method for Data Selection in Mandarin Isolated Word Recognition System

A Combined Task Analysis Method for Data Selection in Mandarin Isolated Word Recognition System

This paper studies the performance of the data selection with a combined task analysis method in task adaptation on Mandarin isolated word recognition. The proposed task analysis method combines coverage unit balanced task analysis with the confusability based analysis. The performance is evaluated with several experiments.

W. Li | J. Wu | Z. G. Wang | Z. Y. He

[1] Ren-Yuan Lyu,et al. Automatic selection of phonetically distributed sentence sets for speaker adaptation with application to large vocabulary Mandarin speech recognition , 1999, Comput. Speech Lang..

[2] Abeer Alwan,et al. Efficient adaptation text design based on the Kullback-Leibler measure , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3] Ron Kohavi,et al. A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[4] Wei Li,et al. An active approach to speaker and task adaptation based on automatic analysis of vocabulary confusability , 2007, INTERSPEECH.

[5] Wei Li,et al. A DTW-based dissimilarity measure for left-to-right hidden Markov models and its application to word confusability analysis , 2006, INTERSPEECH.