Perceive and predict: self-supervised speech representation based loss functions for speech enhancement.
George CloseWilliam RavenscroftThomas HainStefan GoetzePublished in: CoRR (2023)
Keyphrases
- loss function
- speech enhancement
- speech signal
- noise reduction
- noisy environments
- signal to noise ratio
- vocal tract
- single channel
- pairwise
- linear prediction
- loss minimization
- support vector
- speech recognition
- wiener filter
- learning to rank
- pattern recognition
- reproducing kernel hilbert space
- smoothing algorithm
- image restoration
- risk minimization
- training data