Multi-CMGAN+/+: Leveraging Multi-Objective Speech Quality Metric Prediction for Speech Enhancement.
George CloseWilliam RavenscroftThomas HainStefan GoetzePublished in: CoRR (2023)
Keyphrases
- speech enhancement
- multi objective
- linear prediction
- quality metrics
- speech signal
- noise reduction
- noisy environments
- signal to noise ratio
- single channel
- vocal tract
- prediction error
- image coding
- noisy speech
- image quality
- lossless compression
- quality assessment
- background noise
- speech recognition
- wiener filter
- smoothing algorithm
- multi channel
- human perception
- computer vision
- human visual system
- visual quality
- hidden markov models
- wavelet transform
- data analysis
- pattern recognition
- multiscale