"d-o-e-s-not-c-o-m-p-u-t-e": vowel hyperarticulation in speech to an auditory-visual avatar.
Denis BurnhamSebastian JoeffryLauren RicePublished in: AVSP (2010)
Keyphrases
- visual information
- speech signal
- speech synthesis
- visual speech
- vocal tract
- speech recognition
- cross modal
- audio visual
- evoked potentials
- speech recognition systems
- formant frequencies
- prosodic features
- virtual world
- visual features
- signal processing
- speaker independent
- sound source
- speaker identification
- text to speech
- visual data
- fundamental frequency
- human computer interaction
- recognition engine
- speaker recognition
- high level
- eye movements
- multi modal
- visual stimuli
- speech recognizer
- linear prediction
- automatic speech recognition
- non stationary
- spoken language
- information processing
- low level
- hidden markov models
- virtual reality