DocumentCode :
2838696
Title :
Hearer model based stress prediction for Chinese TTS system
Author :
Hu, Guo-Ping ; Liu, Qing-Feng ; Hu, Yu ; Wang, Ren-Hua
Author_Institution :
Univ. of Sci. & Technol. of China, Hefei, China
fYear :
2004
fDate :
15-18 Dec. 2004
Firstpage :
161
Lastpage :
164
Abstract :
People often feel tired if they listen to synthesized speech for a long time. This is mainly because synthesized speech is too flat and never stresses the focus. Unlike traditional TTS research approaches of speaker simulation, the paper investigates stress prediction from the point of view of the hearer. An ideal hearer model is first proposed to predict the stress distribution based on the following hypothesis: people speak with limited stress effort and distribute the limited effort to ensure that the hearer can understand the speaker easily. Then, according to the limited research resource, we modify the ideal hearer model and present a practical model. Experiments show that the stress prediction achieves an acceptable rate of 87.36%.
Keywords :
natural language interfaces; prediction theory; speech processing; speech synthesis; text analysis; Chinese TTS system; hearer model based stress prediction; stress distribution; stress prediction research; synthesized speech; Auditory system; Information analysis; Performance analysis; Predictive models; Rhythm; Speech analysis; Speech processing; Speech synthesis; Stress; Synthesizers;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Chinese Spoken Language Processing, 2004 International Symposium on
Print_ISBN :
0-7803-8678-7
Type :
conf
DOI :
10.1109/CHINSL.2004.1409611
Filename :
1409611
Link To Document :
بازگشت