Phase Reconstruction with Learned Time-Frequency Representations for Single-Channel Speech Separation.
Gordon WichernJonathan Le RouxPublished in: IWAENC (2018)
Keyphrases
- single channel
- speech enhancement
- phase difference
- multi channel
- sound source
- phase unwrapping
- speech signal
- independent component analysis
- audio visual
- speech recognition
- high resolution
- prior information
- frequency domain
- noise reduction
- wavelet decomposition
- brain activity
- image reconstruction
- pattern recognition
- signal to noise ratio
- focus of attention
- noisy environments
- wiener filter
- vocal tract
- single image
- automatic speech recognition
- speech synthesis