# End-to-End Complex-Valued Multidilated Convolutional Neural Network for Joint Acoustic Echo Cancellation and Noise Suppression

arXiv: https://arxiv.org/abs/2110.00745 · HTML: https://mlsanity.com/paper/2110.00745/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 2 Oct 2021.

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [SDR - half-baked or well done?](https://mlsanity.com/paper/1811.02508/) (2018)
- [Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation](https://mlsanity.com/paper/1809.07454/) (2018)
- [DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement](https://mlsanity.com/paper/2008.00264/) (2020)
- [Asteroid: the PyTorch-based audio source separation toolkit for researchers](https://mlsanity.com/paper/2005.04132/) (2020)
- [ACCDOA: Activity-Coupled Cartesian Direction of Arrival Representation for Sound Event Localization and Detection](https://mlsanity.com/paper/2010.15306/) (2020)
- [Densely connected multidilated convolutional networks for dense prediction tasks](https://mlsanity.com/paper/2011.11844/) (2020)
- [Weighted Speech Distortion Losses for Neural-network-based Real-time Speech Enhancement](https://mlsanity.com/paper/2001.10601/) (2020)
- [On Loss Functions for Supervised Monaural Time-Domain Speech Enhancement](https://mlsanity.com/paper/1909.01019/) (2019)
- [The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Speech Quality and Testing Framework](https://mlsanity.com/paper/2001.08662/) (2020)
- [Low-Complexity, Real-Time Joint Neural Echo Control and Speech Enhancement Based On PercepNet](https://mlsanity.com/paper/2102.05245/) (2021)
- [Ensemble of ACCDOA- and EINV2-based Systems with D3Nets and Impulse Response Simulation for Sound Event Localization and Detection](https://mlsanity.com/paper/2106.10806/) (2021)
- [Crowdsourcing approach for subjective evaluation of echo impairment](https://mlsanity.com/paper/2010.13063/) (2020)
- [Y2-Net FCRN for Acoustic Echo and Noise Suppression](https://mlsanity.com/paper/2103.17189/) (2021)
- [Weighted Recursive Least Square Filter and Neural Network based Residual Echo Suppression for the AEC-Challenge](https://mlsanity.com/paper/2102.08551/) (2021)
- [Deep Residual Echo Suppression with A Tunable Tradeoff Between Signal Distortion and Echo Suppression](https://mlsanity.com/paper/2106.13531/) (2021)

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

- [A Generalized Bandsplit Neural Network for Cinematic Audio Source Separation](https://mlsanity.com/paper/2309.02539/) (2023)
- [Rethinking complex-valued deep neural networks for monaural speech enhancement](https://mlsanity.com/paper/2301.04320/) (2023)
- [Deep model with built-in cross-attention alignment for acoustic echo cancellation](https://mlsanity.com/paper/2208.11308/) (2022)
- [Denoising neural networks for magnetic resonance spectroscopy](https://mlsanity.com/paper/2211.00080/) (2022)
- [Room Transfer Function Reconstruction Using Complex-valued Neural Networks and Irregularly Distributed Microphones](https://mlsanity.com/paper/2402.04866/) (2024)
- [Time-Variance Aware Real-Time Speech Enhancement](https://mlsanity.com/paper/2302.13063/) (2023)

## Abstract (arXiv)

Echo and noise suppression is an integral part of a full-duplex communication system. Many recent acoustic echo cancellation (AEC) systems rely on a separate adaptive filtering module for linear echo suppression and a neural module for residual echo suppression. However, not only do adaptive filtering modules require convergence and remain susceptible to changes in acoustic environments, but this two-stage framework also often introduces unnecessary delays to the AEC system when neural modules are already capable of both linear and nonlinear echo suppression. In this paper, we exploit the offset-compensating ability of complex time-frequency masks and propose an end-to-end complex-valued neural network architecture. The building block of the proposed model is a pseudocomplex extension based on the densely-connected multidilated DenseNet (D3Net) building block, resulting in a very small network of only 354K parameters. The architecture utilized the multi-resolution nature of the D3Net building blocks to eliminate the need for pooling, allowing the network to extract features using large receptive fields without any loss of output resolution. We also propose a dual-mask technique for joint echo and noise suppression with simultaneous speech enhancement. Evaluation on both synthetic and real test sets demonstrated promising results across multiple energy-based metrics and perceptual proxies.