BINGO

BraIN imaGined speech cOmmunication
Project ID
Funding Programme:
Basic Research Financing (Horizontal support for all Sciences)
Funding Instrument:
Sub-action 1. Funding New Researchers
Start Date:
28/11/2023
Duration:
24 months
Total Budget:
190,000 EUR
ITI Budget:
190,000 EUR
Scientific Responsible:

Decoding and conveying human thought into the world has been a long-held ambition of the Brain-Computer Interface (BCI) community. Considering that verbal communication via speech is the most natural form of communication, imagined speech is probably the most prominent paradigm to convey the user’s intention in an intuitive way. Despite its great potential, the decoding performance of imagined speech is still far from optimal and significantly lower compared to other BCI paradigms. As a novel BCI paradigm, imagined speech remains understudied while robust feature extraction and classification schemes have not been identified yet. Hence, BINGO’s vision is to create robust algorithms that will decode human speech imagery from EEG signals. The neural processes that underpin imagined speech, will be the cornerstone for the development of neuroinformed decoding schemes, with existing signal processing and machine learning pipelines being explored and tailored accordingly. Additionally, considering that the imagined speech paradigm holds great promise in terms of multiclass scalability, which can lead to high(er) degrees of freedom in a BCI system, BINGO will also explore the potential of incremental vocabulary (i.e., gradually increasing the supported imagined speech vocabulary). Another BINGO aspiration is to investigate the imagined speech at the level of semantic perception (i.e., words of identical meaning expressed in different languages) so as to shed light to the existing neuroscientific hypotheses with respect to the imagined speech paradigm or even give rise to new ones. In this direction, we will). Finally, BINGO’s activities will also be oriented towards increasing its scientific impact and the reproducibility of its results. To this end, a toolbox for decoding speech imagery and a benchmarking framework (i.e., a dataset accompanied by appropriate evaluation metrics) will be made publicly available towards the end of the project’s activities.

Consortium

CENTRE FOR RESEARCH AND TECHNOLOGY HELLAS, Greece
Collaborating Organization(s)
ARISTOTELIO PANEPISTIMIO THESSALONIKIS, Greece

Contact

Dr. Spiridon Nikolopoulos
(Scientific Responsible)
Building A - Office 0.7

Information Technologies Institute
Centre of Research & Technology - Hellas
6th km Harilaou - Thermis, 57001, Thermi - Thessaloniki
Tel.: +30 2311 257752
Fax: +30 2310 474128
Email: nikolopo@iti.gr

Skip to content