Audio-visual speech enhancement using deep neural networks.
Jen-Cheng HouSyu-Siang WangYing-Hui LaiJen-Chun LinYu TsaoHsiu-Wen ChangHsin-Min WangPublished in: APSIPA (2016)
Keyphrases
- audio visual
- speech enhancement
- neural network
- multi modal
- noisy environments
- noise reduction
- signal to noise ratio
- visual information
- single channel
- speech signal
- visual data
- linear prediction
- vocal tract
- pattern recognition
- multi stream
- multimedia
- smoothing algorithm
- speech recognition
- sound source
- audio features
- machine learning
- multiscale
- wiener filter
- brain activity
- metadata
- additive noise
- probabilistic model
- multi channel
- independent component analysis