A corpus-based speech synthesis system with emotion
Speech Communication - Special issue on speech and emotion
Multi-paragraph segmentation of expository text
ACL '94 Proceedings of the 32nd annual meeting on Association for Computational Linguistics
Hi-index | 0.00 |
Representing emotional expressions in text-to-speech synthesis is an interesting subject. The ultimate purpose of our research is to develop an automatic reading system which reads text aloud such as novels with emotion. Our strategy for constructing the system is that we classify the emotion of a text in perspective based on the distribution of emotional words, and classify the emotion of a sentence based on the emotion of nouns, adjectives or verbs composed in the sentence instead of understanding the meaning of the text, and synthesize speech using partially optimum prosodic parameters.