A method for combining intonation modelling and speech unit selection in corpus-based speech synthesis systems.
Francisco Campillo DíazEduardo Rodríguez BangaPublished in: Speech Commun. (2006)
Keyphrases
- speech synthesis
- high accuracy
- high precision
- computational complexity
- error rate
- data sets
- speech recognition
- preprocessing
- pairwise
- significant improvement
- probabilistic model
- distributed systems
- clustering method
- text to speech
- edge detection
- vocal tract
- prosodic features
- fine tuning
- experimental evaluation
- video sequences
- similarity measure
- feature selection