National Changhua University of Education Institutional Repository : Item 987654321/17173
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 6491/11663
造訪人次 : 24647543      線上人數 : 54
RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 進階搜尋


題名: Coping Imbalanced Prosodic Unit Boundary Detection with Linguistically Motivated Prosodic Features
作者: Liu, Yi-Fen;Tseng, Shu-Chuan;Jang, Roger J. S.;Chen, Alvin Cheng-Hsien
貢獻者: 英語學系
關鍵詞: Prosodic unit;Machine learning;Biased minimax probability machine
日期: 2010-09
上傳時間: 2013-08-28T05:01:18Z
出版者: ISCA
摘要: Continuous speech input for ASR processing is usually presegmented into speech stretches by pauses. In this paper, we propose that smaller, prosodically defined units can be identified by tackling the problem on imbalanced prosodic unit boundary detection using five machine learning techniques. A parsimonious set of linguistically motivated prosodic features has been proven to be useful to characterize prosodic boundary information. Furthermore, BMPM is prone to have true positive rate on the minority class, i.e. the defined prosodic units. As a whole, the decision tree classifier, C4.5, reaches a more stable performance than the other algorithms.
關聯: INTERSPEECH 2010, : 1417-1420
顯示於類別:[英語學系] 會議論文


檔案 大小格式瀏覽次數



DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回饋