EmoMix: Emotion Mixing via Diffusion Models for Emotional Speech Synthesis.
Haobin TangXulong ZhangJianzong WangNing ChengJing XiaoPublished in: INTERSPEECH (2023)
Keyphrases
- speech synthesis
- diffusion models
- emotional state
- emotion recognition
- speech recognition
- diffusion model
- affective computing
- information diffusion
- text to speech
- human decision making
- social networks
- affective states
- prosodic features
- vocal tract
- facial expressions
- influence maximization
- viral marketing
- anisotropic diffusion
- hidden markov models
- information retrieval
- audio visual
- multiscale
- bayesian networks