Deep Extractor Network for Target Speaker Recovery From Single Channel Speech Mixtures.
Jun WangJie ChenDan SuLianwu ChenMeng YuYanmin QianDong YuPublished in: CoRR (2018)
Keyphrases
- single channel
- speech enhancement
- speech recognition
- source separation
- multi channel
- automatic speech recognition
- audio visual
- speaker identification
- noise reduction
- vocal tract
- language model
- independent component analysis
- speech signal
- wavelet decomposition
- speaker diarization
- speech synthesis
- computer vision
- motion estimation