The authors recorded whole-brain 3T fMRI from three healthy subjects while they listened to French audio podcasts.
30 hours of fMRI per subject for decoding natural speech.
Cephalonauts One provides deep within-subject fMRI recordings during podcast listening, with a benchmark showing decoding performance improves with more data per subject.
Research Lab
Antoine Collas · Louis Jalouzot · Géraud Ilinca · Corentin Caris · Romain Valabrègue · Ahmed Hassayoune · +8 more
Karavela · UNICOG, CNRS, INSERM, CEA, Paris-Saclay University · LSCP, EHESS, ENS, CNRS, PSL University · Centre de NeuroImagerie de Recherche (CENIR), Sorbonne Université, ICM, Paris, France · Department of Radiology, Hôpital Fondation Adolphe de Rothschild, Paris, France
Research Digest··2 min read
Collas et al.
Why this paper
From Inria, CEA, Paris-Saclay University, Palaiseau, France and 5 others
In one line
Deep per-subject fMRI data improves naturalistic speech decoding performance continuously with more training data.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§