Neuro2Semantic: A Transfer Learning Framework for Semantic Reconstruction of Continuous Language from Human Intracranial EEG
This work addresses a challenge in brain-computer interfaces and neural decoding, offering potential practical applications, though it appears incremental as it builds on existing methods with a novel framework.
The paper tackles the problem of decoding continuous language from intracranial EEG signals by introducing Neuro2Semantic, a transfer learning framework that reconstructs semantic content from neural data, achieving strong performance with only 30 minutes of data and outperforming a state-of-the-art method in low-data settings.
Decoding continuous language from neural signals remains a significant challenge in the intersection of neuroscience and artificial intelligence. We introduce Neuro2Semantic, a novel framework that reconstructs the semantic content of perceived speech from intracranial EEG (iEEG) recordings. Our approach consists of two phases: first, an LSTM-based adapter aligns neural signals with pre-trained text embeddings; second, a corrector module generates continuous, natural text directly from these aligned embeddings. This flexible method overcomes the limitations of previous decoding approaches and enables unconstrained text generation. Neuro2Semantic achieves strong performance with as little as 30 minutes of neural data, outperforming a recent state-of-the-art method in low-data settings. These results highlight the potential for practical applications in brain-computer interfaces and neural decoding technologies.