# MEGConformer: Conformer-Based MEG Decoder for Robust Speech and Phoneme Classification

Source: [MEGConformer: Conformer-Based MEG Decoder for Robust Speech and Phoneme Classification](<https://www.sota2.com/research/paper/692e7311d6b0dd3342932718>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-12T05:00:04.496Z

Authors: Xabier de Zuazo, Ibon Saratxaga, Eva Navas

Published: 2025-12-01T09:25:22.000Z

## Abstract

Decoding speech-related information from non-invasive MEG is a key step toward scalable brain-computer interfaces. We present compact Conformer-based decoders on the LibriBrain 2025 PNPL benchmark for two core tasks: Speech Detection and Phoneme Classification. Our approach adapts a compact Conformer to raw 306-channel MEG signals, with a lightweight convolutional projection layer and task-specific heads. For Speech Detection, a MEG-oriented SpecAugment provided a first exploration of MEG-specific augmentation. For Phoneme Classification, we used inverse-square-root class weighting and a dynamic grouping loader to handle 100-sample averaged examples. In addition, a simple instance-level normalization proved critical to mitigate distribution shifts on the holdout split. Using the official Standard track splits and F1-macro for model selection, our best systems achieved 88.9% (Speech) and 65.8% (Phoneme) on the leaderboard, winning the Phoneme Classification Standard track. For further implementation details, the technical documentation, source code, and checkpoints are available at https://github.com/neural2speech/libribrain-experiments.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Phoneme Classification on LibriBrain (holdout)](<https://www.sota2.com/research/sota/phoneme-classification-on-libribrain-holdout.md>): Phoneme Classification; LibriBrain (holdout); F1 Macro: 65.83.

- [Speech Detection on LibriBrain (holdout)](<https://www.sota2.com/research/sota/speech-detection-on-libribrain-holdout.md>): Speech Detection; LibriBrain (holdout); F1-macro: 88.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.01443>)

[PDF](<https://arxiv.org/pdf/2512.01443.pdf>)

[Implementation](<https://github.com/neural2speech/libribrain-experiments>)
