COMBINING UNIDIRECTIONAL LONG SHORT-TERM MEMORY WITH CONVOLUTIONAL OUTPUT LAYER FOR HIGH-PERFORMANCE SPEECH SYNTHESIS
Wang, Wenfu; Xu, Bo
2017-03
会议名称International Conference on Acoustics, Speech and Signal Processing
页码5500-5504
会议日期2017-3-5
会议地点New Orleans, USA
摘要In this paper, we target improving the accuracy of acoustic modelling for statistical parametric speech synthesis (SPSS) and introduce the convolutional neural network (CNN) due to its powerful capacity in locality modelling. A novel model architecture combining unidirectional long short-term memory (LSTM) and a time-domain convolutional output layer (COL) is proposed and employed to acoustic modelling. The two components complement each other and result in a high-performance synthesis system. Specifically, the unidirectional LSTM can learn expressive feature representations from history context and the COL ingeniously absorbs some of these representations within a look-ahead window to advance predictions. This complementary mechanism significantly improve the predictive accuracy and the quality of synthetic speech. In addition, the unique operation mechanism of convolution makes COL a fine parameter trajectory smoother between consecutive frames. Subjective preference tests show that the proposed architecture can synthesize natural sounding speech without dynamic features.
关键词Statistical Parametric Speech Synthesis Lstm Convolutional Output Layer High-performance Trajectory Smoother
收录类别EI
语种英语
文献类型会议论文
条目标识符http://ir.ia.ac.cn/handle/173211/19660
专题数字内容技术与服务研究中心_听觉模型与认知计算
作者单位Institute of Automation, Chinese Academy of Sciences, Beijing, China
推荐引用方式
GB/T 7714
Wang, Wenfu,Xu, Bo. COMBINING UNIDIRECTIONAL LONG SHORT-TERM MEMORY WITH CONVOLUTIONAL OUTPUT LAYER FOR HIGH-PERFORMANCE SPEECH SYNTHESIS[C],2017:5500-5504.
条目包含的文件 下载所有文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
icassp2017_wang.pdf(228KB)会议论文 开放获取CC BY-NC-SA浏览 下载
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Wang, Wenfu]的文章
[Xu, Bo]的文章
百度学术
百度学术中相似的文章
[Wang, Wenfu]的文章
[Xu, Bo]的文章
必应学术
必应学术中相似的文章
[Wang, Wenfu]的文章
[Xu, Bo]的文章
相关权益政策
暂无数据
收藏/分享
文件名: icassp2017_wang.pdf
格式: Adobe PDF
所有评论 (0)
暂无评论
 

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。