Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme

Abstract Sound source localisation is used in many consumer devices, to isolate audio from individual speakers and reject noise. Localization is frequently accomplished by “beamforming”, which combines phase-shifted audio streams to increase power from chosen source directions, under a known microph...

Full description

Saved in:
Bibliographic Details
Main Authors: Saeid Haghighatshoar, Dylan Richard Muir
Format: Article
Language:English
Published: Nature Portfolio 2025-02-01
Series:Communications Engineering
Online Access:https://doi.org/10.1038/s44172-025-00359-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850067323320270848
author Saeid Haghighatshoar
Dylan Richard Muir
author_facet Saeid Haghighatshoar
Dylan Richard Muir
author_sort Saeid Haghighatshoar
collection DOAJ
description Abstract Sound source localisation is used in many consumer devices, to isolate audio from individual speakers and reject noise. Localization is frequently accomplished by “beamforming”, which combines phase-shifted audio streams to increase power from chosen source directions, under a known microphone array geometry. Dense band-pass filters are often needed to obtain narrowband signal components from wideband audio. These approaches achieve high accuracy, but narrowband beamforming is computationally demanding, and not ideal for low-power IoT devices. We introduce a method for sound source localisation on arbitrary microphone arrays, designed for efficient implementation in ultra-low-power spiking neural networks (SNNs). We use a Hilbert transform to avoid dense band-pass filters, and introduce an event-based encoding method that captures the phase of the complex analytic signal. Our approach achieves high accuracy for SNN methods, comparable with traditional non-SNN super-resolution beamforming. We deploy our method to low-power SNN inference hardware, with much lower power consumption than super-resolution methods. We demonstrate that signal processing approaches co-designed with spiking neural network implementations can achieve much improved power efficiency. Our Hilbert-transform-based method for beamforming can also improve the efficiency of traditional digital signal processing.
format Article
id doaj-art-536efdf597654f788f367fe676e1ab5e
institution DOAJ
issn 2731-3395
language English
publishDate 2025-02-01
publisher Nature Portfolio
record_format Article
series Communications Engineering
spelling doaj-art-536efdf597654f788f367fe676e1ab5e2025-08-20T02:48:22ZengNature PortfolioCommunications Engineering2731-33952025-02-014111010.1038/s44172-025-00359-9Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding schemeSaeid Haghighatshoar0Dylan Richard Muir1SynSenseSynSenseAbstract Sound source localisation is used in many consumer devices, to isolate audio from individual speakers and reject noise. Localization is frequently accomplished by “beamforming”, which combines phase-shifted audio streams to increase power from chosen source directions, under a known microphone array geometry. Dense band-pass filters are often needed to obtain narrowband signal components from wideband audio. These approaches achieve high accuracy, but narrowband beamforming is computationally demanding, and not ideal for low-power IoT devices. We introduce a method for sound source localisation on arbitrary microphone arrays, designed for efficient implementation in ultra-low-power spiking neural networks (SNNs). We use a Hilbert transform to avoid dense band-pass filters, and introduce an event-based encoding method that captures the phase of the complex analytic signal. Our approach achieves high accuracy for SNN methods, comparable with traditional non-SNN super-resolution beamforming. We deploy our method to low-power SNN inference hardware, with much lower power consumption than super-resolution methods. We demonstrate that signal processing approaches co-designed with spiking neural network implementations can achieve much improved power efficiency. Our Hilbert-transform-based method for beamforming can also improve the efficiency of traditional digital signal processing.https://doi.org/10.1038/s44172-025-00359-9
spellingShingle Saeid Haghighatshoar
Dylan Richard Muir
Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
Communications Engineering
title Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
title_full Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
title_fullStr Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
title_full_unstemmed Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
title_short Low-power Spiking Neural Network audio source localisation using a Hilbert Transform audio event encoding scheme
title_sort low power spiking neural network audio source localisation using a hilbert transform audio event encoding scheme
url https://doi.org/10.1038/s44172-025-00359-9
work_keys_str_mv AT saeidhaghighatshoar lowpowerspikingneuralnetworkaudiosourcelocalisationusingahilberttransformaudioeventencodingscheme
AT dylanrichardmuir lowpowerspikingneuralnetworkaudiosourcelocalisationusingahilberttransformaudioeventencodingscheme