English Speech Prosody Modeling Improvement and its Application to TTS
碩士 === 國立交通大學 === 電信工程研究所 === 107 === This paper proposes to use special word as leading word to define the various type of word-chunks base on the conventional syntactic tree structure generated by the Stanford Parser, and improves the affecting factors of the prosody modeling. The model is trained...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2019
|
Online Access: | http://ndltd.ncl.edu.tw/handle/svk98a |