Audio-visual Speech Enhancement Using Conditional Variational Auto-Encoder.
Mostafa SadeghiSimon LeglaiveXavier Alameda-PinedaLaurent GirinRadu HoraudPublished in: CoRR (2019)
Keyphrases
- audio visual
- speech enhancement
- noisy environments
- noise reduction
- multi modal
- signal to noise ratio
- single channel
- speech signal
- bit rate
- visual information
- linear prediction
- vocal tract
- multi stream
- multimedia
- sound source
- image segmentation
- wiener filter
- visual data
- smoothing algorithm
- multi channel
- audio features
- edge detection
- optical flow
- eye movements
- machine learning