Continuous decoding of brain activity to text using EEG
Recently, real-time electroencephalographic (EEG) understanding has emerged as a prominent tool for human-computer interaction and healthcare. EEG provides a key to the brain-computer interface (BCI) framework to translate human cognition patterns to interact with machines without any physical intervention. However, despite the EEG being extensively employed in both clinical settings and brain-computer interface (BCI), real-world EEG-based applications are not yet well established. EEG signals fluctuate rapidly and are subject to various sources of noise and artifacts including environmental noise, frequency interference etc. Thus, the key issue concerning an EEG-based BCI system is to interpret EEG signals to accurately understand the user’s intent.
To bring the power of EEG monitoring of neural activity out of the clinic, we propose an EEG-based continuous brain-to-text decoding (B2TD) system using novel energy-efficient deep learning algorithms. This proposal will build an intracortical B2TD system that decodes imagined handwriting movements from neural activity (in the motor cortex) or understanding from visual reading into summarized texts.
EEG-based Brain-to-Text Decoding (B2TD) Framework
