Efficient reduction of Gaussian components using MDL criterion for HMM-based speech recognition
暂无分享,去创建一个
[1] 篠田 浩一. A study on robustness against data insufficiency for speech recognition , 2001 .
[2] J. Fritsch. Mixture trees - hierarchically tied mixture densities for modeling HMM emission probabilities , 1999, EUROSPEECH.
[3] Man-Hung Siu,et al. Pruning of state-tying tree using bayesian information criterion with multiple mixtures , 2000, INTERSPEECH.
[4] Ananth Sankar,et al. Parameter tying and gaussian clustering for faster, better, and smaller speech recognition , 1999, EUROSPEECH.
[5] Mei-Yuh Hwang,et al. Dynamically configurable acoustic models for speech recognition , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).
[6] Ramesh A. Gopinath,et al. Model selection in acoustic modeling , 1999, EUROSPEECH.
[7] Jj Urgen Fritsch. Mixture Trees Hierarchically Tied Mixture Densities for Modeling Hmm Emission Probabilities , .
[8] Koichi Shinoda,et al. MDL-based context-dependent subword modeling for speech recognition , 2000 .
[9] Yonghong Yan,et al. Dynamic threshold setting via Bayesian information criterion (BIC) in HMM training , 2000, INTERSPEECH.
[10] Koichi Shinoda,et al. Speech recognition using tree-structured probability density function , 1994, ICSLP.
[11] Jorma Rissanen,et al. Universal coding, information, prediction, and estimation , 1984, IEEE Trans. Inf. Theory.