Prosodic cues for emotion characterization in real-life spoken dialogs.
Laurence DevillersIoana VasilescuPublished in: INTERSPEECH (2003)
Keyphrases
- real life
- text to speech synthesis
- speech recognition
- prosodic features
- multimodal fusion
- text to speech
- facial expressions
- dialog systems
- speech synthesis
- emotion recognition
- real world
- virtual humans
- mixed initiative
- spoken language
- speaker verification
- application domains
- language model
- speech retrieval
- virtual agents
- automatic speech recognition
- language understanding
- high robustness
- visual cues
- speech signal
- hidden markov models