首页>
外国专利>
CONTEXT DEPENDENT HUMAN LIKE EMOTIONAL SPEECH SYNTHESIS IN YOUR VOICE WITH PROSODY
CONTEXT DEPENDENT HUMAN LIKE EMOTIONAL SPEECH SYNTHESIS IN YOUR VOICE WITH PROSODY
展开▼
机译:带有语音的依赖于上下文的人类情感语音合成
展开▼
页面导航
摘要
著录项
相似文献
摘要
The invention relates to an innovative technique for synthesizing human like emotional speech in any person™s voice with added prosody effects by just recording some 900 spoken words in person™s voice considering the context for the phonemes. An object of the invention is to provide a way to synthesize human like speech in the voice of any person who does the voice recording required. A further object of the invention is to improve the quality of synthesized speech so that emotions like anger, happiness, sadness, fear and surprize are felt in the synthesized voice. A still further object of the invention is to provide a facility for time gap adjustments, pitch contour and energy contour adjustments for improving the naturalness in the synthesized speech. The different prosodic features namely the effect of question mark, exclamatory mark and all punctuation marks can be added in the synthesized speech generated using pitch contour adjustments and time gap adjustments.
展开▼