Neuralink discloses direct thought-to-audio speech decoding in ALS participant Terry, the VOICE trial's third implantee
Neuralink published a video this week showing Terry, a married man with bulbar-onset ALS, saying 'Hello, how are you doing today?' and 'I, I love you' to his wife through an implanted N1 brain-computer interface. The output routed through Grok Voice, xAI's speech engine, using a clone of Terry's own pre-ALS recordings. It is Neuralink's most detailed publicly disclosed thought-to-speech demonstration to date and the first to route through xAI's Grok Voice engine for personalised voice synthesis. Another VOICE participant, Kenneth Shock, appeared in a Neuralink clip on 16 September 2026, two days before Terry's disclosure, saying 'I love you' to his wife. Terry is the third participant in Neuralink's VOICE early-feasibility trial (ClinicalTrials.gov NCT07224256), formally titled 'VOICE: An Early Feasibility Study of a Precise Robotically Implanted Brain-Computer Interface for Communication Restoration.' VOICE is a separate protocol from Neuralink's PRIME motor-decoding trial and its CAN-PRIME and GB-PRIME international arms. The N1 implant holds an FDA Breakthrough Device Designation; it is not FDA-approved and not commercially available. Prior Neuralink participants Noland Arbaugh (PRIME P1), Alex (PRIME P2), and Brad Smith (PRIME first ALS participant) are on the motor decoding track. Brad Smith's earlier speech restoration used cursor-typed text through a cloned voice; Terry's demonstration is the first Neuralink disclosure of direct neural-to-audio decoding. Neuralink has not published words-per-minute, word-error-rate, latency, or vocabulary-size metrics. No peer-reviewed methods paper accompanies the demonstration.