WESPER: Zero-shot and Realtime Whisper to Normal Voice Conversion for Whisper-based Speech Interactions.
Jun RekimotoPublished in: CHI (2023)
Keyphrases
- text to speech
- real time
- emotion recognition
- speech synthesis
- speech recognition
- speech signal
- speech recognition errors
- speech quality
- speech sounds
- voice activity detection
- automatic speech recognition
- dialogue system
- fundamental frequency
- prosodic features
- spoken dialogue systems
- endpoint detection
- synthesized speech
- automatic speech recognition systems
- spoken document retrieval
- speaker recognition
- graphics hardware
- linear prediction
- real time systems
- multi modal
- probabilistic model
- natural language
- pattern recognition
- social networks