Interaural coherence induced ideal binary mask for binaural speech separation and dereverberation.
Yi-Ting ChenTzu-Hao ChenMao-Chang HuangTai-Shih ChiPublished in: ISCSLP (2016)
Keyphrases
- sound source
- speech signal
- audio visual
- speech recognition
- audio features
- speaker identification
- automatic speech recognition
- vocal tract
- noisy environments
- automatic speech recognition systems
- hidden markov models
- linear prediction
- single channel
- focus of attention
- blind source separation
- non stationary
- multi modal
- endpoint detection
- emotion recognition
- non binary
- speaker verification
- speaker recognition
- multi stream
- real time