Concss: Contrastive-based Context Comprehension for Dialogue-Appropriate Prosody in Conversational Speech Synthesis.
Yayue DengJinlong XueYukang JiaQifei LiYichen HanFengping WangYingming GaoDengfeng KeYa LiPublished in: ICASSP (2024)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- prosodic features
- vocal tract
- speech corpus
- conversational agent
- contextual knowledge
- contextual information
- hidden markov models
- conversational agents
- context sensitive
- multi modal
- spoken language
- speech acts
- dialogue system
- context dependent
- context aware
- feature selection