Towards a segmental vocoder driven by ultrasound and optical images of the tongue and lips.
Thomas HueberGérard CholletBruce DenbyGérard DreyfusMaureen StonePublished in: INTERSPEECH (2008)
Keyphrases
- optical images
- ultrasound images
- satellite images
- video images
- imaging systems
- high resolution
- synthetic aperture radar
- sar images
- range data
- hidden markov models
- surface model
- active contour model
- image data
- shortest path
- human faces
- edge detection
- active contours
- camera calibration
- d objects
- multi sensor
- three dimensional