DNN driven Speaker Independent Audio-Visual Mask Estimation for Speech Separation.
Mandar GogateAhsan AdeelRicard MarxerJon BarkerAmir HussainPublished in: CoRR (2018)
Keyphrases
- audio visual
- digit recognition
- speaker independent
- multi modal
- speech recognition
- visual information
- speaker dependent
- multi stream
- visual data
- audio features
- emotion recognition
- sound source
- speaker verification
- multimedia
- hidden markov models
- speaker identification
- audio visual speech recognition
- speech recognizer
- n gram
- computer vision
- information retrieval