Researchers at the University of Texas at Austin have developed a groundbreaking method for decoding language from non-invasive brain recordings, using functional magnetic resonance imaging (fMRI). Their approach, detailed in their 2023 study published in Nature Neuroscience, focuses on reconstructing spoken and imagined language through a sophisticated decoder trained on extensive fMRI data. The system achieved significant results in translating brain activity into coherent word sequences, demonstrating its potential for aiding individuals with communication disorders. This interview with Jerry Tang, a lead researcher, explores the goals, methodology, and future directions of their work, including the promise of translating their approach to more portable technologies like functional near-infrared spectroscopy (fNIRS).