High-quality speech synthesis for phonetic speech segmentation.
Fabrice MalfrèreThierry DutoitPublished in: EUROSPEECH (1997)
Keyphrases
- speech synthesis
- speech recognition
- high quality
- vocal tract
- text to speech
- speech signal
- prosodic features
- segmentation method
- language model
- speech recognizer
- speech corpus
- hidden markov models
- segmentation algorithm
- image segmentation
- automatic speech recognition
- level set
- speaker identification
- segmentation accuracy
- speech recognition systems
- speech sounds
- image analysis
- pattern recognition
- acoustic models
- medical images
- multiscale
- object segmentation
- shape prior
- region growing
- energy function
- speaker adaptation
- noisy environments
- fully automatic
- ground truth
- computer vision