Tag
The paper introduces LibriBrain100, a large-scale MEG dataset with over 100 hours of data for neural speech decoding, achieving state-of-the-art performance on word classification benchmarks.