SICRN: Advancing Speech Enhancement through State Space Model and Inplace Convolution Techniques.
Changjiang ZhaoShulin HeXueliang ZhangPublished in: ICASSP (2024)
Keyphrases
- state space model
- speech enhancement
- noise reduction
- noisy environments
- signal to noise ratio
- additive noise
- linear prediction
- autoregressive
- single channel
- speech signal
- kalman filter
- state estimation
- vocal tract
- image processing
- wiener filter
- multi channel
- kalman filtering
- smoothing algorithm
- noisy images
- maximum likelihood
- pattern recognition
- high resolution
- multiscale