Search papers, labs, and topics across Lattice.
Cephalonauts One provides a novel, deep whole-brain fMRI dataset (30 hours per subject) of naturalistic speech comprehension, enabling unprecedented investigation into neural representations of auditory language processing. The dataset facilitates a new brain decoding benchmark focused on audio segment retrieval from fMRI activity, allowing researchers to assess model performance in reconstructing speech content from neural signals. A key finding is the continuous improvement in decoding accuracy with increased training data, suggesting that larger datasets will unlock more precise neural decoding of speech.
A single 30-hour fMRI dataset unlocks whole-brain decoding of naturalistic speech, revealing continuous performance gains with more data.
Cephalonauts One is a whole-brain 3 Tesla (3T) functional magnetic resonance imaging (fMRI) dataset recorded while subjects listened to audio podcasts. Three healthy subjects underwent multiple scanning sessions, each consisting of five 15-minute runs, while listening to podcasts in their native language. With 30 hours of fMRI data per subject, the current release is the deepest available fMRI dataset using naturalistic speech stimuli. The dataset pairs brain activity with the corresponding podcast audio, transcript annotations, and derived stimulus embeddings. Furthermore, we introduce a brain decoding benchmark formulated as audio segment retrieval: given fMRI activity from a held-out session, the decoder must identify the corresponding time-aligned podcast audio segment among candidate segments. We provide standardized splits, evaluation metrics, and baseline decoders for this task. Finally, a scaling analysis shows that decoding performance improves continuously with the amount of training data per subject.
