Dong Ming, Liu Run-sheng. Transcendental Information Based Triphone Model Tying Structure Adaptation Strategy[J]. Journal of Electronics & Information Technology, 2007, 29(9): 2050-2053. doi: 10.3724/SP.J.1146.2006.00200
Citation:
Dong Ming, Liu Run-sheng. Transcendental Information Based Triphone Model Tying Structure Adaptation Strategy[J]. Journal of Electronics & Information Technology, 2007, 29(9): 2050-2053. doi: 10.3724/SP.J.1146.2006.00200
Dong Ming, Liu Run-sheng. Transcendental Information Based Triphone Model Tying Structure Adaptation Strategy[J]. Journal of Electronics & Information Technology, 2007, 29(9): 2050-2053. doi: 10.3724/SP.J.1146.2006.00200
Citation:
Dong Ming, Liu Run-sheng. Transcendental Information Based Triphone Model Tying Structure Adaptation Strategy[J]. Journal of Electronics & Information Technology, 2007, 29(9): 2050-2053. doi: 10.3724/SP.J.1146.2006.00200
A Transcendental Information Based (TIB) triphone model tying structure adaptation strategy is delivered, and this strategy can improve the triphone model tying structure to suit the target co-pronunciation features with small amount of adaptation data. The TIB triphone model tying structure adaptation strategy uses the baseline acoustic models triphone model tying result as the transcendental model clustering center, with the adaptation data alignment by the baseline acoustic model, re-estimate the TIB triphone model clustering center and model tying structure recursively under maximum likelihood principle. The experiments show that the TIB triphone model tying structure adaptation strategy can improve the triphone model tying structure with only 2 hours adaptation corpus, and in the experiment of English acoustic model for Chinese speakers, the TIB strategy will increase the recognition accuracy rate from 74.59% to 83.63%.
Lee K F and Hon H W. Speaker-independent phone recognition using hidden Markov models[J].IEEE Trans on ASSP.1989, 37(11):1641-1648[2]Chang E, Shi Y, Zhou J L, and Huang C. Speech lab in a box: A Mandarin speech toolbox to jumpstart speech related research. Eurospeech 2001, Aalborg, Denmark, 2001.[3]Young S and Evermann G, et al.. The HTK Book (for HTK Version 3.2). Cambridge University Engineering Department, 2002.[4]Lazarides A, Normandin Y, and Kuhn R. Improving decision trees for acoustic modeling. Proceedings of ICSLP'96. Philadelphia, 1996: 1053-1056.[5]Liang W Q, Liu J, and Liu R S. An automatic pronunciation quality assessing algorithm for computer assisted language learning. Chinese Journal of Electronics, 2005, 14(4):639-643.