Bypassing Direct Reconstruction: Speech Detection from MEG via Large-Scale Audio Retrieval

TL;DR

This paper introduces a novel two-step framework for decoding speech from MEG signals by retrieving matching audio segments from large databases and directly detecting speech, achieving top performance in a benchmark task.

Contribution

The authors propose a new approach that bypasses direct reconstruction by combining contrastive audio retrieval with speech detection, improving accuracy in MEG-based speech decoding.

Findings

Achieved first place in the LibriBrain 2025 Speech Detection task with an F1-score of 0.962.

Leveraging external audio databases significantly enhances speech detection from MEG signals.

Abstract

Decoding speech from non-invasive brain signals is challenging. For the LibriBrain 2025 Speech Detection task, we propose a novel two-step framework that bypasses direct reconstruction. First, a contrastive learning model retrieves the matching speech segment for the given test MEG from a large-scale audio library (LibriVox). Second, a speech detection model generates the binary silence/speech sequence directly from this retrieved audio. With this approach, our team Sherlock Holmes achieved first place in the extended track (F1-score: 0.962), demonstrating that leveraging external audio databases is a highly effective strategy.