Speech perception and production are uniquely human traits, which inherently limit the scope of neuroscience tools available to study them. While most human research methodologies are limited to non-invasive techniques such as EEG (Electroencephalography) and fMRI (functional Magnetic Resonance Imaging) our lab focuses on collaborative research with rare neurosurgical patients undergoing treatment for refractory epilepsy. These patients are implanted with intracranial electrode grids placed directly on the surface of the brain for a one-week period in order to monitor and localize epileptogenic activity. While the electrode placement is driven solely by clinical necessity, during lulls in clinical care there is a unique opportunity to conduct cognitive tasks with patients in the hospital while Electrocorticographic (ECoG) signals are being collected directly from cortex. These neural signals provide an aggregate measure of neural excitability with combined temporal and spatial resolution. The focus of the research in the lab is to employ various experimental, machine learning and signal processing approaches to elucidate speech networks based on neural signals recorded directly from human cortex. Our goal is to further our understanding of how the human cortex supports speech perception and production while providing better tools for clinicians to map function prior to surgery.