Corpus ID: 235266003

Multi-Scale Attention Neural Network for Acoustic Echo Cancellation

  title={Multi-Scale Attention Neural Network for Acoustic Echo Cancellation},
  author={Lu Ma and Song Yang and Yaguang Gong and Zhongqin Wu},
Acoustic Echo Cancellation (AEC) plays a key role in speech interaction by suppressing the echo received at microphone introduced by acoustic reverberations from loudspeakers. Since the performance of linear adaptive filter (AF) would degrade severely due to nonlinear distortions, background noises, and microphone clipping in real scenarios, deep learning has been employed for AEC for its good nonlinear modelling ability. In this paper, we constructed an end-to-end multi-scale attention neural… Expand

Figures and Tables from this paper

Joint AEC AND Beamforming with Double-Talk Detection using RNN-Transformer
  • Vinay Kothapally, Yong Xu, Meng Yu, Shi-Xiong Zhang, Dong Yu
  • Computer Science, Engineering
  • ArXiv
  • 2021
An all-deep-learning framework that combines multi-channel AEC and the recently proposed self-attentive recurrent neural network (RNN) beamformer and a double-talk detection transformer (DTDT) module based on the multi-head attention transformer structure that computes attention over time by leveraging frame-wise double- talk predictions is proposed. Expand


Deep Learning for Acoustic Echo Cancellation in Noisy and Double-Talk Scenarios
Experimental results show the effectiveness of the proposed method for echo removal in double-talk, background noise, and nonlinear distortion scenarios and it can be generalized to untrained speakers. Expand
Deep Multitask Acoustic Echo Cancellation
Experimental results show that the proposed deep learning based method outperforms the existing methods for unseen speakers in terms of the echo return loss enhancement (ERLE) for single-talk periods and the perceptual evaluation of speech quality (PESQ) score for double- talk periods. Expand
Acoustic Echo Cancellation by Combining Adaptive Digital Filter and Recurrent Neural Network
A fusion scheme by combining adaptive filter and neural network is proposed for AEC, validating the effectiveness and superiority of the proposed combination scheme. Expand
CAD-AEC: Context-Aware Deep Acoustic Echo Cancellation
Experiments show that the proposed CAD-AEC can robustly achieve better echo return loss enhancement (ERLE) and perceptual speech quality compared to the previous classical and deep learning techniques. Expand
Conv-TasNet: Surpassing Ideal Time–Frequency Magnitude Masking for Speech Separation
  • Yi Luo, N. Mesgarani
  • Computer Science, Medicine
  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • 2019
A fully convolutional time-domain audio separation network (Conv-TasNet), a deep learning framework for end-to-end time- domain speech separation, which significantly outperforms previous time–frequency masking methods in separating two- and three-speaker mixtures. Expand
An Attention-based Neural Network Approach for Single Channel Speech Enhancement
Experiments show that the proposed attention approach can consistently achieve better performance in terms of speech quality (PESQ) and intelligibility (STOI). Expand
Suppressing acoustic echo in a spectral envelope space
This paper proposes an echo suppression algorithm, which estimates the spectral envelope of the echo signal by spectral modification-a technique originally proposed for noise reduction, and shows that this new approach has several advantages over the traditional AEC. Expand
Adaptation of a memoryless preprocessor for nonlinear acoustic echo cancelling
Experiments under adverse conditions and with real hardware demonstrate robust convergence with both models, and an echo reduction improvement by up to 10 dB at amplitude peaks. Expand
Acoustic Echo Control
This chapter recalls the established signal processing strategies for acoustic echo cancellation, mostly based on the family of deterministic least-squares adaptive filters, and presents different optimum filtering solutions which were tailored for robustness against the specific uncertainties related to echo path variability, nonlinearity, and clock drift in a system. Expand
Nonlinear residual acoustic echo suppression for high levels of harmonic distortion
This work presents a new AEC architecture that consists of a linear, subband adaptive AEC filter that followed a nonlinear residual echo suppression (RES) stage specifically designed to address harmonic distortion. Expand