Multi-View Audio-Articulatory Features for Phonetic Recognition on RTMRI-TIMIT Database.
Ioannis K. DourosAthanasios KatsamanisPetros MaragosPublished in: ICASSP (2018)
Keyphrases
- multi view
- single view
- feature extraction
- multiple views
- speech recognition
- camera calibration
- d objects
- depth map
- three dimensional
- hidden markov models
- range images
- object recognition
- view synthesis
- semi supervised
- feature set
- face alignment
- feature vectors
- multi view clustering
- multi view face detection
- co training
- multi views
- multiple viewpoints
- visual information
- feature space
- action recognition