How auditory and visual prosody is used in end-of-utterance detection.
Pashiera BarkhuysenEmiel KrahmerMarc SwertsPublished in: INTERSPEECH (2006)
Keyphrases
- visual information
- evoked potentials
- thermal images
- visual features
- cross modal
- detection rate
- false positives
- object detection
- detection algorithm
- automatic detection
- information processing
- signal processing
- real time
- anomaly detection
- speech recognition
- low level
- hidden markov models
- human visual system
- detection accuracy
- pattern recognition
- spoken language
- image processing
- data sets