Acoustic measures vs. phonetic features as predictors of audible discontinuity in concatenative speech synthesis.
Hisashi KawaiMinoru TsuzakiPublished in: INTERSPEECH (2002)
Keyphrases
- speech synthesis
- speech recognition
- emotional speech
- feature vectors
- speech recognition systems
- feature extraction
- classification accuracy
- co occurrence
- emotion recognition
- benchmark datasets
- false positives
- prosodic features
- speaker independent
- spatial information
- vocal tract
- statistical measures
- neural network
- image processing