mlsanityData to 7 Oct 2026

Phase-Aware Deep Speech Enhancement: It's All About The Frame Length

arXiv 2203.16222 ↗First posted 30 Mar 2022eess.AS cs.LG cs.SDcited by 5

Not on the map. Why?

Abstract (arXiv)

Algorithmic latency in speech processing is dominated by the frame length used for Fourier analysis, which in turn limits the achievable performance of magnitude-centric approaches. As previous studies suggest the importance of phase grows with decreasing frame length, this work presents a systematical study on the contribution of phase and magnitude in modern Deep Neural Network (DNN)-based speech enhancement at different frame lengths. Results indicate that DNNs can successfully estimate phase when using short frames, with similar or better overall performance compared to using longer frames. Thus, interestingly, modern phase-aware DNNs allow for low-latency speech enhancement at high quality.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 16 of its 40 references that are in the map. First posted on arXiv 30 Mar 2022.

Builds on

16 of 40 references.

References · 16

  1. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,320
  2. 2018SDR - half-baked or well done?cited by 420
  3. 2018Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separationcited by 400
  4. 2020Attention is All You Need in Speech Separationcited by 184
  5. 2019Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separationcited by 182
  6. 2020DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancementcited by 133
  7. 2020The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Testing Framework, and Challenge Resultscited by 128
  8. 2018The Conversation: Deep Audio-Visual Speech Enhancementcited by 107
  9. 2019Filterbank design for end-to-end speech separationcited by 16
  10. 2019Demystifying TasNet: A Dissecting Approachcited by 12
  11. 2021On The Compensation Between Magnitude and Phase in Speech Separationcited by 12
  12. 2019A Multi-Phase Gammatone Filterbank for Speech Separation via TasNetcited by 11
4 more
  1. 2020Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networkscited by 10
  2. 2018Phasebook and Friends: Leveraging Discrete Representations for Source Separationcited by 5
  3. 2020Phase reconstruction based on recurrent phase unwrapping with deep neural networkscited by 4
  4. 2016Phase reconstruction of spectrograms with linear unwrapping: application to audio signal restorationcited by 1

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 1; 2024: 0; 2025: 2.

5 papers.

  1. 2022Speech Enhancement and Dereverberation with Diffusion-based Generative Modelscited by 195
  2. 2025A2SB: Audio-to-Audio Schrodinger Bridgescited by 11
  3. 2022Efficient Transformer-based Speech Enhancement Using Long Frames and STFT Magnitudescited by 10
  4. 2023Enhanced Neural Beamformer with Spatial Information for Target Speech Extractioncited by 0
  5. 2025Hybrid Real- and Complex-valued Neural Network Architecturecited by 0