mlsanityData to 14 Oct 2026

From KAN to GR-KAN: Advancing Speech Enhancement with KAN-Based Methodology

arXiv 2412.17778 ↗First posted 23 Dec 2024eess.AS cs.AI cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

Deep neural network (DNN)-based speech enhancement (SE) usually uses conventional activation functions, which lack the expressiveness to capture complex multiscale structures needed for high-fidelity SE. Group-Rational KAN (GR-KAN), a variant of Kolmogorov-Arnold Networks (KAN), retains KAN's expressiveness while improving scalability on complex tasks. We adapt GR-KAN to existing DNN-based SE by replacing dense layers with GR-KAN layers in the time-frequency (T-F) domain MP-SENet and adapting GR-KAN's activations into the 1D CNN layers in the time-domain Demucs. Results on Voicebank-DEMAND show that GR-KAN requires up to 4x fewer parameters while improving PESQ by up to 0.1. In contrast, KAN, facing scalability issues, outperforms MLP on a small-scale signal modeling task but fails to improve MP-SENet. We demonstrate the first successful use of KAN-based methods for consistent improvement in both time- and SoTA TF-domain SE, establishing GR-KAN as a promising alternative for SE.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 22 of its 39 references that are in the map. First posted on arXiv 23 Dec 2024.

Builds on

22 of 39 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  3. 2016Gaussian Error Linear Units (GELUs)cited by 3,502
  4. 2024KAN: Kolmogorov-Arnold Networkscited by 649
  5. 2020Real Time Speech Enhancement in the Waveform Domaincited by 186
  6. 2017SEGAN: Speech Enhancement Generative Adversarial Networkcited by 166
  7. 2021MetricGAN+: An Improved Version of MetricGAN for Speech Enhancementcited by 114
  8. 2019MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancementcited by 104
  9. 2024KAN or MLP: A Fairer Comparisoncited by 73
  10. 2019Pad'e Activation Units: End-to-end Learning of Flexible Activation Functions in Deep Networkscited by 67
  11. 2014Learning Activation Functions to Improve Deep Neural Networkscited by 63
  12. 2024Mamba in Speech: Towards an Alternative to Self-Attentioncited by 60
10 more
  1. 2024Kolmogorov-Arnold Transformercited by 56
  2. 2022CMGAN: Conformer-based Metric GAN for Speech Enhancementcited by 45
  3. 2023SELM: Speech Enhancement Using Discrete Tokens and Language Modelscited by 36
  4. 2021DPT-FSNet: Dual-path Transformer Based Full-band and Sub-band Fusion Network for Speech Enhancementcited by 33
  5. 2020Review and Comparison of Commonly Used Activation Functions for Deep Neural Networkscited by 21
  6. 2023Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancementcited by 13
  7. 2022SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Taskscited by 8
  8. 2024SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Modelcited by 6
  9. 2022TridentSE: Guiding Speech Enhancement with 32 Global Tokenscited by 4
  10. 2024Selective State Space Model for Monaural Speech Enhancementcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025Aligning Generative Speech Enhancement with Perceptual Feedbackcited by 3
  2. 2026Training-Free Intelligibility-Guided Observation Addition for Noisy ASRcited by 0