Dynamical predictive coding with reservoir computing performs noise-robust multi-sensory speech recognition
Frontiers in Computational Neuroscience · 23.09.2024 · 18
DOI: 10.3389/fncom.2024.1464603 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Computational Neuroscience, online erschienen am 23.09.2024. Verfasst von Yoshihiro Yonemura und 1 weiteren Autor:innen. Beteiligte Einrichtung: Graduate of System Information Science, Future University Hakodate, Hakodate, Hokkaido, Japan. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Multi-sensory integration is a perceptual process through which the brain synthesizes a unified perception by integrating inputs from multiple sensory modalities. A key issue is understanding how the brain performs multi-sensory integrations using a common neural basis in the cortex. A cortical model based on reservoir computing has been proposed to elucidate the role of recurrent connectivity among cortical neurons in this process. Reservoir computing is well-suited for time series processing, such as speech recognition. This inquiry focuses on extending a reservoir computing-based cortical model to encompass multi-sensory integration within the cortex. This research introduces a dynamical model of multi-sensory speech recognition, leveraging predictive coding combined with reservoir computing. Predictive coding offers a framework for the hierarchical structure of the cortex. The model integrates reliability weighting, derived from the computational theory of multi-sensory integration, to adapt to multi-sensory time series processing. The model addresses a multi-sensory speech recognition task, necessitating the management of complex time series. We observed that the reservoir effectively recognizes speech by extracting time-contextual information and weighting sensory inputs according to sensory noise. These findings indicate that the dynamic properties of recurrent networks are applicable to multi-sensory time series processing, positioning reservoir computing as a suitable model for multi-sensory integration.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Graduate of System Information Science, Future University Hakodate, Hakodate, Hokkaido, Japan.
- International Research Center for Neurointelligence (IRCN), The University of Tokyo, Tokyo, Japan.
Zitieren
APA
Yonemura, Y., & Katori, Y. (2024). Dynamical predictive coding with reservoir computing performs noise-robust multi-sensory speech recognition. Frontiers in Computational Neuroscience, 18. https://doi.org/10.3389/fncom.2024.1464603
Vancouver
Yonemura Y, Katori Y. Dynamical predictive coding with reservoir computing performs noise-robust multi-sensory speech recognition. Frontiers in Computational Neuroscience. 2024;18. doi:10.3389/fncom.2024.1464603
BibTeX
@article{Yonemura2024,
author = {Yonemura, Yoshihiro and Katori, Yuichi},
title = {Dynamical predictive coding with reservoir computing performs noise-robust multi-sensory speech recognition},
journal = {Frontiers in Computational Neuroscience},
year = {2024},
volume = {18},
doi = {10.3389/fncom.2024.1464603}
}
EUVIMED