Multi-Channel Speech Enhancement Using Time-Domain Convolutional Denoising Autoencoder.
Naohiro TawaraTetsunori KobayashiTetsuji OgawaPublished in: INTERSPEECH (2019)
Keyphrases
- single channel
- speech enhancement
- denoising
- frequency domain
- multi channel
- restricted boltzmann machine
- wiener filter
- image denoising
- spatial domain
- natural images
- image processing
- noisy images
- wavelet domain
- signal to noise ratio
- conditional random fields
- speech signal
- noisy environments
- independent component analysis
- computer vision