Metraipolis › archive › dc_34e3bb28c500

Zephyr: An Efficient Audio Denoising System Using Spiking Neural Networks Enabled With A Sparsity-Aware Flexible FPGA PE Array — arXiv:2609.37711

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37711
License (as archived)
CC0-1.0
Archived
2026-10-07T16:59:19.217Z
SHA-256 of the archived text
a959854697e8c083c824feef6ef269544b4876433f083f3c2559c360d959583b
Stable URL of this page
https://metraipolis.com/documento/dc_34e3bb28c500

How to cite

Zephyr: An Efficient Audio Denoising System Using Spiking Neural Networks Enabled With A Sparsity-Aware Flexible FPGA PE Array — arXiv:2609.37711 — arXiv (en), https://arxiv.org/abs/2609.37711 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_34e3bb28c500 (Metraipolis archive dc_34e3bb28c500, sha256 a959854697e8c083c824feef6ef269544b4876433f083f3c2559c360d959583b)
BibTeX
@misc{dc34e3bb28c500,
  title = {Zephyr: An Efficient Audio Denoising System Using Spiking Neural Networks Enabled With A Sparsity-Aware Flexible FPGA PE Array — arXiv:2609.37711},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37711},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_34e3bb28c500; sha256 a959854697e8c083c824feef6ef269544b4876433f083f3c2559c360d959583b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37711. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In this work we look to neuromorphic computing to solve the power consumption problem that audio denoising neural networks face on edge devices like smartphones, wireless headphones and hearing aids. Spiking neural networks (SNNs) have the potential to solve this problem due to their high activation sparsity and low complexity, however many SOTA SNNs require hardware that supports a mixture of operations to be able to fully perform inference. To solve this problem, we convert SOTA audio denoising neural network Spiking-FullSubNet to a hardware friendly version showing that via QAT and activation function simplification we can achieve $\approx28\times$ improvement in power consumption to 52.9nJ per 32ms audio frame when calculated for custom digital hardware in a 45nm process node. We then propose a digital circuit which by means of a sparsity-aware flexible PE array can perform inference of the heterogeneous compute load of Spiking-FullSubNet, and validate this circuit on a PYNQ-Z1 FPGA achieving a real-time factor of 0.727 at 100MHz.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AR).
Categorie arXiv: Hardware Architecture (cs.AR) ; Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.37711 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T16:31:49.773Z dalla pagina https://arxiv.org/catchup/cs.AR/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).