# Complex-Valued Time-Frequency Self-Attention for Speech Dereverberation

arXiv: https://arxiv.org/abs/2211.12632 · HTML: https://mlsanity.com/paper/2211.12632/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 33 references that are in the map.
- First posted on arXiv 22 Nov 2022.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [Self-Attention Generative Adversarial Networks](https://mlsanity.com/paper/1805.08318/) (2018)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [On the Relationship between Self-Attention and Convolutional Layers](https://mlsanity.com/paper/1911.03584/) (2019)
- [Deep Complex Networks](https://mlsanity.com/paper/1705.09792/) (2017)
- [Data augmentation and loss normalization for deep noise suppression](https://mlsanity.com/paper/2008.06412/) (2020)
- [Speech Dereverberation Using Fully Convolutional Networks](https://mlsanity.com/paper/1803.08243/) (2018)
- [SkipConvNet: Skip Convolutional Neural Network for Speech Dereverberation using Optimally Smoothed Spectral Mapping](https://mlsanity.com/paper/2007.09131/) (2020)
- [Deep Speech Enhancement for Reverberated and Noisy Signals using Wide Residual Networks](https://mlsanity.com/paper/1901.00660/) (2019)
- [Speech Dereverberation Based on Integrated Deep and Ensemble Learning Algorithm](https://mlsanity.com/paper/1801.04052/) (2018)

## Built on by (most cited first, 6 of 6)

- [AV-RIR: Audio-Visual Room Impulse Response Estimation](https://mlsanity.com/paper/2312.00834/) (2023)
- [AdVerb: Visually Guided Audio Dereverberation](https://mlsanity.com/paper/2308.12370/) (2023)
- [Deep neural network techniques for monaural speech enhancement: state of the art analysis](https://mlsanity.com/paper/2212.00369/) (2022)
- [A High-Fidelity Speech Super Resolution Network using a Complex Global Attention Module with Spectro-Temporal Loss](https://mlsanity.com/paper/2507.00229/) (2025)
- [Speech Separation based on Contrastive Learning and Deep Modularization](https://mlsanity.com/paper/2305.10652/) (2023)
- [Emerging Approaches for THz Array Imaging: A Tutorial Review and Software Tool](https://mlsanity.com/paper/2309.08844/) (2023)

## Abstract (arXiv)

Several speech processing systems have demonstrated considerable performance improvements when deep complex neural networks (DCNN) are coupled with self-attention (SA) networks. However, the majority of DCNN-based studies on speech dereverberation that employ self-attention do not explicitly account for the inter-dependencies between real and imaginary features when computing attention. In this study, we propose a complex-valued T-F attention (TFA) module that models spectral and temporal dependencies by computing two-dimensional attention maps across time and frequency dimensions. We validate the effectiveness of our proposed complex-valued TFA module with the deep complex convolutional recurrent network (DCCRN) using the REVERB challenge corpus. Experimental findings indicate that integrating our complex-TFA module with DCCRN improves overall speech quality and performance of back-end speech applications, such as automatic speech recognition, compared to earlier approaches for self-attention.