Discriminatively trained recurrent neural networks for single-channel speech separation.
Felix WeningerJohn R. HersheyJonathan Le RouxBjörn W. SchullerPublished in: GlobalSIP (2014)
Keyphrases
- recurrent neural networks
- single channel
- discriminatively trained
- speech enhancement
- sound source
- multi channel
- object detection
- speech signal
- sequence classification
- audio visual
- discriminative learning
- discriminative training
- neural network
- feed forward
- independent component analysis
- generative model
- automatic speech recognition
- echo state networks
- frequency domain
- wavelet decomposition
- speech recognition
- hidden markov models
- wiener filter
- prior information
- artificial neural networks
- brain activity
- signal to noise ratio
- signal processing
- multi modal
- noise reduction
- training data
- higher order
- pattern recognition
- decision making