Neural Speech Enhancement with Very Low Algorithmic Latency and Complexity via Integrated Full- and Sub-Band Modeling.
Zhong-Qiu WangSamuele CornellShukjae ChoiYounglo LeeByeong-Yeol KimShinji WatanabePublished in: CoRR (2023)
Keyphrases
- subband
- speech enhancement
- filter bank
- linear prediction
- multiresolution
- discrete wavelet transform
- wavelet transform
- high frequency
- wavelet coefficients
- feature vectors
- wavelet decomposition
- frequency domain
- bit rate
- noisy environments
- computational complexity
- wavelet domain
- motion estimation
- speech signal
- frequency band
- single channel