MEGState: Phoneme Decoding from Magnetoencephalography Signals
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918256677224448 |
|---|---|
| author | Suzuki, Shuntaro Hsu, Chia-Chun Dan Tsao, Yu Sugiura, Komei |
| author_facet | Suzuki, Shuntaro Hsu, Chia-Chun Dan Tsao, Yu Sugiura, Komei |
| contents | Decoding linguistically meaningful representations from non-invasive neural recordings remains a central challenge in neural speech decoding. Among available neuroimaging modalities, magnetoencephalography (MEG) provides a safe and repeatable means of mapping speech-related cortical dynamics, yet its low signal-to-noise ratio and high temporal dimensionality continue to hinder robust decoding. In this work, we introduce MEGState, a novel architecture for phoneme decoding from MEG signals that captures fine-grained cortical responses evoked by auditory stimuli. Extensive experiments on the LibriBrain dataset demonstrate that MEGState consistently surpasses baseline model across multiple evaluation metrics. These findings highlight the potential of MEG-based phoneme decoding as a scalable pathway toward non-invasive brain-computer interfaces for speech. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_17978 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MEGState: Phoneme Decoding from Magnetoencephalography Signals Suzuki, Shuntaro Hsu, Chia-Chun Dan Tsao, Yu Sugiura, Komei Neurons and Cognition Machine Learning Sound Decoding linguistically meaningful representations from non-invasive neural recordings remains a central challenge in neural speech decoding. Among available neuroimaging modalities, magnetoencephalography (MEG) provides a safe and repeatable means of mapping speech-related cortical dynamics, yet its low signal-to-noise ratio and high temporal dimensionality continue to hinder robust decoding. In this work, we introduce MEGState, a novel architecture for phoneme decoding from MEG signals that captures fine-grained cortical responses evoked by auditory stimuli. Extensive experiments on the LibriBrain dataset demonstrate that MEGState consistently surpasses baseline model across multiple evaluation metrics. These findings highlight the potential of MEG-based phoneme decoding as a scalable pathway toward non-invasive brain-computer interfaces for speech. |
| title | MEGState: Phoneme Decoding from Magnetoencephalography Signals |
| topic | Neurons and Cognition Machine Learning Sound |
| url | https://arxiv.org/abs/2512.17978 |