mlsanityData to 7 Oct 2026

DeepFilterNet2: Towards Real-Time Speech Enhancement on Embedded Devices for Full-Band Audio

arXiv 2205.05474 ↗First posted 11 May 2022eess.AS cs.LG cs.SDcited by 23

Not on the map. Why?

Abstract (arXiv)

Deep learning-based speech enhancement has seen huge improvements and recently also expanded to full band audio (48 kHz). However, many approaches have a rather high computational complexity and require big temporal buffers for real time usage e.g. due to temporal convolutions or attention. Both make those approaches not feasible on embedded devices. This work further extends DeepFilterNet, which exploits harmonic structure of speech allowing for efficient speech enhancement (SE). Several optimizations in the training procedure, data augmentation, and network structure result in state-of-the-art SE performance while reducing the real-time factor to 0.04 on a notebook Core-i5 CPU. This makes the algorithm applicable to run on embedded devices in real-time. The DeepFilterNet framework can be obtained under an open source license.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 15 of its 22 references that are in the map. First posted on arXiv 11 May 2022.

Builds on

15 of 22 references.

References · 15

  1. 2017Don't Decay the Learning Rate, Increase the Batch Sizecited by 313
  2. 2021DNSMOS P.835: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressorscited by 165
  3. 2020DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancementcited by 133
  4. 2021Interspeech 2021 Deep Noise Suppression Challengecited by 43
  5. 2021DeepFilterNet: A Low Complexity Speech Enhancement Framework for Full-Band Audio based on Deep Filteringcited by 38
  6. 2017A Hybrid DSP/Deep Learning Approach to Real-Time Full-Band Speech Enhancementcited by 34
  7. 2021Towards efficient models for real-time deep noise suppressioncited by 28
  8. 2022FullSubNet+: Channel Attention FullSubNet with Complex Spectrograms for Speech Enhancementcited by 28
  9. 2021Glance and Gaze: A Collaborative Learning Framework for Single-channel Speech Enhancementcited by 25
  10. 2020A Perceptually-Motivated Approach for Low-Complexity, Real-Time Enhancement of Fullband Speechcited by 23
  11. 2021DCCRN+: Channel-wise Subband DCCRN with SNR Estimation for Speech Enhancementcited by 22
  12. 2021Real-time Denoising and Dereverberation with Tiny Recurrent U-Netcited by 16
3 more
  1. 2019Deep Filtering: Signal Extraction and Reconstruction Using Complex Time-Frequency Filterscited by 14
  2. 2021S-DCCRN: Super Wide Band DCCRN with learnable complex feature for speech enhancementcited by 12
  3. 2020CLCNet: Deep learning-based Noise Reduction for Hearing Aids using Complex Linear Codingcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 0; 2025: 11; 2026: 7.

23 papers.

Sort:

By area:

  1. 2023DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancementcited by 25
  2. 2023Ultra Low Complexity Deep Learning Based Noise Suppressioncited by 7
  3. 2023A Mask Free Neural Network for Monaural Speech Enhancementcited by 3
  4. 2023ROSE: A Recognition-Oriented Speech Enhancement Framework in Air Traffic Control Using Multi-Objective Learningcited by 2
  5. 2025DeepFilterGAN: A Full-band Real-time Speech Enhancement System with GAN-based Stochastic Regenerationcited by 1
  6. 2025TF-MLPNet: Tiny Real-Time Neural Speech Separationcited by 1
  7. 2025Leveraging Discriminative Latent Representations for Conditioning GAN-Based Speech Enhancementcited by 1
  8. 2025Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patternscited by 1
  9. 2025Proactive Hearing Assistants that Isolate Egocentric Conversationscited by 1
  10. 2023Continuous Modeling of the Denoising Process for Speech Enhancement Based on Deep Learningcited by 0
  11. 2025DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Aidscited by 0
  12. 2025UL-UNAS: Ultra-Lightweight U-Nets for Real-Time Speech Enhancement via Network Architecture Searchcited by 0
11 more
  1. 2025How much to Dereverberate? Low-Latency Single-Channel Speech Enhancement in Distant Microphone Scenarioscited by 0
  2. 2025Model as Loss: A Self-Consistent Training Paradigmcited by 0
  3. 2025Revealing the Role of Audio Channels in ASR Performance Degradationcited by 0
  4. 2025IS3 : Generic Impulsive--Stationary Sound Separation in Acoustic Scenes using Deep Filteringcited by 0
  5. 2026An Interpretable, Controllable Time-Varying IIR Denoiser for On-Device Assistive Hearingcited by 0
  6. 2026Visual-Informed Speech Enhancement Using Attention-Based Beamformingcited by 0
  7. 2026QC-GAN: A Parameter-Efficient Quaternion Conformer GAN for High-Fidelity Speech Enhancementcited by 0
  8. 2026Cloud-Boosted Low-Compute Multi-Channel Speech Enhancementcited by 0
  9. 2026RT-SEMamba: Real-Time Speech Enhancement Mamba via Progressive Knowledge Distillationcited by 0
  10. 2026SlimDiffuSE: Towards Efficient Diffusion-Based Speech Enhancement using Slimmable Networkscited by 0
  11. 2026VietPrism: A large-scale Vietnamese speech and deepfake corpus with diverse dialects and code-switchingcited by 0