Joint separation and denoising of noisy multi-talker speech using recurrent neural networks and permutation invariant training.
Morten KolbaekDong YuZheng-Hua TanJesper JensenPublished in: MLSP (2017)
Keyphrases
- recurrent neural networks
- denoising
- recurrent networks
- echo state networks
- feedforward neural networks
- feed forward
- neural network
- translation invariant
- noisy environments
- image denoising
- feed forward neural networks
- complex valued
- hearing impaired
- artificial neural networks
- speech recognition
- reservoir computing
- total variation
- hidden layer
- cascade correlation
- speech signal
- image processing
- training algorithm
- nonlinear dynamic systems
- computer vision
- natural images
- neural model
- noisy images
- automatic speech recognition
- sound source
- chaotic time series
- denoising methods
- real time