Abstract : We consider a P300 BCI application where the subjects can write figures and letters in an unsupervised fashion. We (i) show that a generic speller can attain the state-of-the-art accuracy without any training phase or calibration and (ii) present an adaptive setup that consistently increases the bit rate for most of the subjects.
https://hal.inria.fr/hal-01104024
Contributor : Maureen Clerc
<>
Submitted on : Saturday, January 17, 2015 - 8:50:15 PM
Last modification on : Friday, February 22, 2019 - 1:24:00 AM
Long-term archiving on : Friday, September 11, 2015 - 6:53:16 AM
Emmanuel Daucé, Eoin Thomas. Evidence build-up facilitates on-line adaptivity in dynamic environments: example of the BCI P300-speller. 22nd European Symposium on Artificial Neural Networks, Apr 2014, Bruges, Belgium. ⟨hal-01104024⟩
