Weighted Speech Distortion Losses for Neural-network-based Real-time Speech Enhancement.
Yangyang XiaSebastian BraunChandan K. A. ReddyHarishchandra DubeyRoss CutlerIvan TashevPublished in: CoRR (2020)
Keyphrases
- speech enhancement
- real time
- noisy environments
- noise reduction
- speech signal
- signal to noise ratio
- vocal tract
- linear prediction
- single channel
- noisy speech
- speech recognition
- neural network
- wiener filter
- smoothing algorithm
- background noise
- additive noise
- automatic speech recognition
- edge preserving
- speech synthesis
- image coding
- edge detection
- multiscale