LA-VOCE: LOW-SNR Audio-Visual Speech Enhancement Using Neural Vocoders.
Rodrigo MiraBuye XuJacob DonleyAnurag KumarStavros PetridisVamsi Krishna IthapuMaja PanticPublished in: ICASSP (2023)
Keyphrases
- audio visual
- low snr
- speech enhancement
- noisy environments
- multi modal
- noise reduction
- signal to noise ratio
- single channel
- speech signal
- noise model
- visual information
- visual data
- neural network
- vocal tract
- linear prediction
- multi channel
- sound source
- low signal to noise ratio
- wiener filter
- multimedia
- speech recognition
- edge detection
- human motion
- gaussian noise
- image data
- pattern recognition
- video sequences
- keywords