Micro-ECoG pseudoword speech repetition

This enclosed data is collected using high-density, high channel-count micro-electrocorticography (micro-ECoG) neural recordings from awake neurosurgical patients in the operating room. Arrays used were either 256-channel grids (1.72 mm electrode spacing), or 128-channel strips (1.33 mm electrode spacing). Patients repeated pseudowords that were either one syllable (consonant-vowel-consonant) or two syllables (vowel-consonant-vowel), from a total inventory of 9 different phonemes. Task duration was approximately 10-15 minutes. The data were collected at the Duke University Medical Center and approved by the Institutional Review Board of the Duke University Health System (Pro00072892). This dataset is curated by Drs. Gregory Cogan and Jonathan Viventi and is formatted in the BIDS standard with continuous data that contains both auditory markings and response markings for tokens and phonemes. No audio data is uploaded for HIPAA compliance. Anatomical locations of each electrode are also included.