Audiovisual synthesis of exaggerated speech for corrective feedback in computer-assisted pronunciation training.
Junhong ZhaoHua YuanWai-Kim LeungHelen M. MengJia LiuShanhong XiaPublished in: ICASSP (2013)
Keyphrases
- computer assisted
- speech recognition
- surgical training
- acoustic models
- audio visual
- computer aided
- intraoperative
- automatic speech recognition
- hearing impaired
- hidden markov models
- spontaneous speech
- emotion recognition
- foreign language
- training process
- speech signal
- pattern recognition
- automatic speech recognition systems
- online learning
- training set
- dialogue system
- multimedia content
- visual information
- broadcast news
- speech recognizer
- speaker independent
- facial expressions
- image analysis
- image processing
- grapheme to phoneme conversion