Speech-driven lip motion generation with a trajectory HMM.
Gregor HoferJunichi YamagishiHiroshi ShimodairaPublished in: INTERSPEECH (2008)
Keyphrases
- speech recognition
- hidden markov models
- speech signal
- visual speech recognition
- multi stream
- visual speech
- automatic speech recognition
- automatic speech recognition systems
- space time
- optical flow
- motion estimation
- image sequences
- facial motion
- robot motion
- configuration space
- motion trajectories
- motion planning
- constant velocity
- audio visual
- speech recognizer
- motion model
- motion features
- motion capture
- speaker identification
- static images
- moving points
- moving objects
- humanoid robot
- human motion
- motion segmentation
- noisy environments
- motion analysis
- keyword spotting
- recognition engine
- object trajectories
- motion patterns
- dynamic textures
- vocal tract
- kinematic model
- visual data
- speaker independent
- camera motion
- spatio temporal
- angular velocity
- mouth region
- handwriting recognition