QQ客服 官方微博 反馈留言

	Perceptual evaluation weight training for text-to-speech synthesis
	Wu Zhiyong ; Cai Lianhong ; Cai Rui
	2010-05-06 ; 2010-05-06
关键词	Practical/ optimisation speech synthesis text analysis/ perceptual evaluation weight training text-to-speech synthesis perceptual listening test supervised weight training optimization user-behavior feedback/ B6130E Speech recognition and synthesis B0260 Optimisation techniques C5260S Speech processing techniques C6130D Document processing techniques C1180 Optimisation techniques
中文摘要	A weight training approach was developed based on perceptual evaluations for unit selection in concatenative text-to-speech (TTS) synthesis. The speech units are selected based on current weights, with the synthesized results evaluated syllable by syllable by a perceptual listening test. The result gives new candidate units with perceptual scores. Then the weights are modified to match the new candidates to provide supervised weight training. Tests demonstrate that the method improves the naturalness of the synthesized result from about 3.49 to 4.02. The method also allows automatic optimization of the TTS system according to the user-behavior feedback.
语种	英语 ; 英语
出版者	Tsinghua Univ. Press ; China
内容类型	期刊论文
源URL	[http://hdl.handle.net/123456789/9845]
专题	清华大学
推荐引用方式 GB/T 7714	Wu Zhiyong,Cai Lianhong,Cai Rui. Perceptual evaluation weight training for text-to-speech synthesis[J],2010, 2010.
APA	Wu Zhiyong,Cai Lianhong,&Cai Rui.(2010).Perceptual evaluation weight training for text-to-speech synthesis..
MLA	Wu Zhiyong,et al."Perceptual evaluation weight training for text-to-speech synthesis".(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。