Emodiff: Intensity Controllable Emotional Text-to-Speech with Soft-Label Guidance.
Yiwei GuoChenpeng DuXie ChenKai YuPublished in: ICASSP (2023)
Keyphrases
- text to speech
- speech synthesis
- programming tool
- word processing
- prosodic features
- text to speech synthesis
- image intensity
- intensity values
- multi label
- english text
- class labels
- emotional state
- using artificial neural networks
- distance learning
- image labeling
- multi modal
- emotion recognition
- pattern recognition
- writing skills
- training set