A single channel speech enhancement approach by combining statistical criterion and multi-frame sparse dictionary learning.
Hung-Wei TsengSrikanth VishnubhotlaMingyi HongXiangfeng WangJinjun XiaoZhi-Quan LuoTao ZhangPublished in: INTERSPEECH (2013)
Keyphrases
- dictionary learning
- single channel
- speech enhancement
- multi frame
- wiener filter
- sparse representation
- multi channel
- motion estimation
- image sequences
- point correspondences
- super resolution
- computationally efficient
- independent component analysis
- motion analysis
- optical flow
- signal to noise ratio
- filtering algorithm
- frequency domain
- signal processing
- noisy environments
- noise reduction
- optic flow
- prior information
- feature points
- speech signal
- video sequences
- feature selection
- face recognition
- brain activity
- input image
- image patches
- computer vision
- image restoration