Hybrid EEG--Driven Brain--Computer Interface: A Large Language Model Framework for Personalized Language Rehabilitation

TL;DR

This paper introduces a hybrid EEG-driven BCI framework utilizing large language models to personalize and adapt language rehabilitation for users with speech and motor impairments in real time.

Contribution

It presents a novel system combining EEG-based BCIs with LLMs to enable personalized, adaptive language therapy for neurological patients.

Findings

Demonstrated real-time neural intent detection using EEG.

Enabled personalized language exercises based on neural signals.

Adjusted task difficulty dynamically through neural effort monitoring.

Abstract

Conventional augmentative and alternative communication (AAC) systems and language-learning platforms often fail to adapt in real time to the user's cognitive and linguistic needs, especially in neurological conditions such as post-stroke aphasia or amyotrophic lateral sclerosis. Recent advances in noninvasive electroencephalography (EEG)--based brain-computer interfaces (BCIs) and transformer--based large language models (LLMs) offer complementary strengths: BCIs capture users' neural intent with low fatigue, while LLMs generate contextually tailored language content. We propose and evaluate a novel hybrid framework that leverages real-time EEG signals to drive an LLM-powered language rehabilitation assistant. This system aims to: (1) enable users with severe speech or motor impairments to navigate language-learning modules via mental commands; (2) dynamically personalize vocabulary, sentence-construction exercises, and corrective feedback; and (3) monitor neural markers of cognitive effort to adjust task difficulty on the fly.