QQ客服 官方微博 反馈留言

	Automatic phonetic segmentation using HMM model
	Wang Li-Juan ; Cao Zhi-Gang
	2010-05-06 ; 2010-05-06
关键词	Practical Theoretical or Mathematical Experimental/ cepstral analysis Gaussian processes hidden Markov models regression analysis speech processing speech recognition trees (mathematics)/ automatic phonetic segmentation HMM model automatic speech recognition system text-to-speech segmentation forced alignment mode optimal acoustic feature selection static 12D Mel-frequency cepstral coefficient Gaussian mixture components regression tree speaker-dependent tri-phone HMM models acoustic unit boundary/ B6130E Speech recognition and synthesis B0240J Markov processes B0250 Combinatorial mathematics C5260S Speech processing techniques C1250C Speech recognition C1140J Markov processes C1160 Combinatorial mathematics
中文摘要	HMM models are widely used in the automatic speech recognition system to segment text-to-speech (TTS) units in the forced alignment mode. To improve the segmentation performance, the optimal acoustic feature selection and the training condition of the HMM model are discussed. Experimental results show that the static 12-D Mel-frequency cepstral coefficient (MFCC) feature is the optimal acoustic feature; the optimal number of Gaussian mixture components per state is 1; the optimal number of tied states after model clustering by the classification and regression tree (CART) is about 3000 for speaker-dependent tri-phone HMM models. With optimized parameters, the segmentation accuracy on English test corpus is increased from 77.3% to 85.4%.
语种	中文 ; 中文
出版者	Nanjing Univ. of Aeronautics & Astronautics ; China
内容类型	期刊论文
源URL	[http://hdl.handle.net/123456789/11076]
专题	清华大学
推荐引用方式 GB/T 7714	Wang Li-Juan,Cao Zhi-Gang. Automatic phonetic segmentation using HMM model[J],2010, 2010.
APA	Wang Li-Juan,&Cao Zhi-Gang.(2010).Automatic phonetic segmentation using HMM model..
MLA	Wang Li-Juan,et al."Automatic phonetic segmentation using HMM model".(2010).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。