Description of Work
T3.1 – Data Collection Protocol (M6-M14)
Aims to formulate the data collection protocol. The protocol will be defined based on both the knowledge gained as part of T2.1 and literature review, including among others the provided cue/stimulus, the type and number of the imagined words/phonemes and the words that will be explored as part of the activities of T2.4. Additionally, the protocol must be formulated in such a way that it will facilitate the investigation towards an incremental vocabulary as well as cross-session testing. Finally, synchronization mechanisms, like LSL, will be explored, aiming to record the available data streams (i.e., cue/stimulus, EEG, timestamps) in a synchronous fashion.
T3.2 – Data Recording (M10-M21)
This task materializes the data collection protocol(s) developed in T3.1 and includes the recording of the EEG data for at least 30 participants. Following the standard practice of EEG research in social science and humanities involving working with human participants, the written informed consent will be obtained for all participants, and their personal data will be privately stored. Finally, the whole process will be oriented towards neuroprotection (i.e. the explicit protection of neurorights) based on the practices identified by T1.3.
T3.3 – Evaluation Plan (M15-M24)
This task is dedicated to the evaluation of the BINGO’s developed algorithms, with respect to the imagined speech decoding, incremental vocabulary learning and interconnection between languages. In this direction, an early prototype will be developed in order to support the online system evaluation. The evaluation will be based upon both existing measurements (e.g. classification accuracy, ITR, etc.) and on metrics identified as part of this task, appropriately tailored to imagined speech decoding, that will allow us to establish a benchmarking framework. Finally, a comparison with existing SoTA imagined speech decoding algorithms will also be performed.