Speaker independent audio-visual continuous speech recognition.
Luhong LiangXiaoxing LiuYibao ZhaoXiaobo PiAra V. NefianPublished in: ICME (2) (2002)
Keyphrases
- continuous speech recognition
- audio visual
- speaker independent
- digit recognition
- speech recognition
- hidden markov models
- multi modal
- speaker identification
- multi stream
- visual information
- audio features
- visual data
- multimedia
- speaker verification
- speech recognition systems
- speech recognizer
- emotion recognition
- broadcast news
- high dimensional
- gaussian mixture model
- action recognition
- knn
- training set