Next-generation neural interface: ALS patient transmits nearly 2 million words through thought alone
Imagine a world where paralysis does not rob you of your voice. This is no longer science fiction. Researchers at the University of California, Davis, have achieved a breakthrough, demonstrating how a patient with amyotrophic lateral sclerosis (ALS) was able to communicate fully at home using a speech neuroprosthesis. Over 19 months, Casey Harrell transmitted 183,060 sentences — that's 1,960,163 words — at an average speed of 56 words per minute. The results were published in the prestigious journal Nature Medicine.
A home system that works without scientists
Unlike most laboratory demonstrations, this study focuses on real, everyday life. Harrell used the system for over 3,800 hours without the constant presence of researchers. He communicated with family, friends, colleagues, and doctors, sent messages, participated in video calls, and even maintained full-time employment. According to the patient's own assessment, 92% of sentences were decoded as at least "mostly correct." In formal tests, accuracy exceeded 99% with a vocabulary of 125,000 English words, reaching a peak of 99.2%.
How it works: from thought to text and voice
In 2023, Harrell had four microelectrode arrays implanted in his left precentral gyrus — the brain region responsible for speech coordination. The system reads signals from 256 cortical electrodes when the patient attempts to speak. An algorithm converts neural activity into phoneme probabilities every 80 milliseconds, and a language model selects the most likely sequence of words. The text is displayed on a screen in real-time and can then be vocalized using a synthesized voice, tuned to Harrell's voice before his illness. Notably, the same cursor decoder, operating through the speech motor cortex, is used to control the computer, refuting previous assumptions that different brain areas are needed for speech and movement.
From lab to everyday life: a key step forward
This study marks a transition from controlled trials to practical use. The authors emphasize that 3,800 hours of recorded brain activity constitute the largest individual dataset with single-neuron resolution. However, it is important to understand that the system remains experimental. It requires wired connections, daily setup by trained assistants, and due to its size, is only suitable for home use. It is not yet known how well the results can be transferred to other patients with different implant locations and electrode types.
My expert opinion: This case is undoubtedly a historic step for neuroprosthetics. It proves that the technology can move beyond laboratories and truly change people's lives. However, the path to widespread adoption is still long. The main challenges are miniaturization, wireless data transmission, and adapting the system to each patient's individual neuroanatomy. Nevertheless, 2 million words transmitted by the power of thought is a powerful signal that the future has already arrived.