MsEmoTTS: Multi-scale emotion transfer, prediction, and control for emotional speech synthesis.
Yi LeiShan YangXinsheng WangLei XiePublished in: CoRR (2022)
Keyphrases
- speech synthesis
- multiscale
- speech recognition
- text to speech
- prediction accuracy
- control system
- emotional state
- emotion recognition
- vocal tract
- affective computing
- prediction algorithm
- multiple scales
- prosodic features
- pattern recognition
- human behaviour
- image processing
- cognitive model
- facial expressions
- knowledge transfer
- virtual characters
- control method
- optimal control
- prediction model
- human computer interaction
- data mining