Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks

Understanding cognitive processes in the brain demands sophisticated models capable of replicating neural dynamics at large scales. We present a physiologically inspired speech recognition architecture, compatible and scalable with deep learning frameworks, and demonstrate that end-to-end gradient d...

Full description

Saved in:
Bibliographic Details
Main Authors: Alexandre Bittar, Philip N. Garner
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-09-01
Series:Frontiers in Neuroscience
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fnins.2024.1449181/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850264130605285376
author Alexandre Bittar
Alexandre Bittar
Philip N. Garner
author_facet Alexandre Bittar
Alexandre Bittar
Philip N. Garner
author_sort Alexandre Bittar
collection DOAJ
description Understanding cognitive processes in the brain demands sophisticated models capable of replicating neural dynamics at large scales. We present a physiologically inspired speech recognition architecture, compatible and scalable with deep learning frameworks, and demonstrate that end-to-end gradient descent training leads to the emergence of neural oscillations in the central spiking neural network. Significant cross-frequency couplings, indicative of these oscillations, are measured within and across network layers during speech processing, whereas no such interactions are observed when handling background noise inputs. Furthermore, our findings highlight the crucial inhibitory role of feedback mechanisms, such as spike frequency adaptation and recurrent connections, in regulating and synchronizing neural activity to improve recognition performance. Overall, on top of developing our understanding of synchronization phenomena notably observed in the human auditory pathway, our architecture exhibits dynamic and efficient information processing, with relevance to neuromorphic technology.
format Article
id doaj-art-e05d8260d3284aaaafc1e4ee3897ddde
institution OA Journals
issn 1662-453X
language English
publishDate 2024-09-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neuroscience
spelling doaj-art-e05d8260d3284aaaafc1e4ee3897ddde2025-08-20T01:54:46ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2024-09-011810.3389/fnins.2024.14491811449181Exploring neural oscillations during speech perception via surrogate gradient spiking neural networksAlexandre Bittar0Alexandre Bittar1Philip N. Garner2Idiap Research Institute, Audio Inference, Martigny, SwitzerlandÉcole Polytechnique Fédérale de Lausanne, Lausanne, SwitzerlandÉcole Polytechnique Fédérale de Lausanne, Lausanne, SwitzerlandUnderstanding cognitive processes in the brain demands sophisticated models capable of replicating neural dynamics at large scales. We present a physiologically inspired speech recognition architecture, compatible and scalable with deep learning frameworks, and demonstrate that end-to-end gradient descent training leads to the emergence of neural oscillations in the central spiking neural network. Significant cross-frequency couplings, indicative of these oscillations, are measured within and across network layers during speech processing, whereas no such interactions are observed when handling background noise inputs. Furthermore, our findings highlight the crucial inhibitory role of feedback mechanisms, such as spike frequency adaptation and recurrent connections, in regulating and synchronizing neural activity to improve recognition performance. Overall, on top of developing our understanding of synchronization phenomena notably observed in the human auditory pathway, our architecture exhibits dynamic and efficient information processing, with relevance to neuromorphic technology.https://www.frontiersin.org/articles/10.3389/fnins.2024.1449181/fullneural oscillationsspiking neural networksspeech recognitionbrain-inspired computingdeep learningsurrogate gradient
spellingShingle Alexandre Bittar
Alexandre Bittar
Philip N. Garner
Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
Frontiers in Neuroscience
neural oscillations
spiking neural networks
speech recognition
brain-inspired computing
deep learning
surrogate gradient
title Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
title_full Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
title_fullStr Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
title_full_unstemmed Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
title_short Exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
title_sort exploring neural oscillations during speech perception via surrogate gradient spiking neural networks
topic neural oscillations
spiking neural networks
speech recognition
brain-inspired computing
deep learning
surrogate gradient
url https://www.frontiersin.org/articles/10.3389/fnins.2024.1449181/full
work_keys_str_mv AT alexandrebittar exploringneuraloscillationsduringspeechperceptionviasurrogategradientspikingneuralnetworks
AT alexandrebittar exploringneuraloscillationsduringspeechperceptionviasurrogategradientspikingneuralnetworks
AT philipngarner exploringneuraloscillationsduringspeechperceptionviasurrogategradientspikingneuralnetworks