Voice of the Future: Neurointerfaces for Amyotrophic Lateral Sclerosis Communication

Amyotrophic lateral sclerosis (ALS) is a degenerative neurological condition that causes muscle weakness. In advanced stages, it progresses to tetraplegia accompanied by loss of speech and inability to communicate.[1] Brain–computer interfaces (BCIs) can help individuals with ALS by restoring their ability to speak. Such devices work by decoding intracortical brain signals into intended speech.

For this, in 2024, a clinical demonstration was conducted in which a 45-year-old man with advanced-stage ALS and severe dysarthria was surgically implanted with four microelectrode arrays. Just 25 days later, this system showed promising results. On the first day, testing showed that the system achieved 99.6% accuracy with a limited vocabulary and 90.2% on the second day with a much larger vocabulary. This shows that with little training, a patient will be able to communicate like other normal people.[2]

However, these text-based BCIs miss important aspects of human speech, such as tone, rhythm, and the inability to hear one's voice simultaneously. For this, in 2025, scientists implanted 256 microelectrodes in the ventral precentral gyrus of a man with ALS who can no longer speak. This system could instantly decode his neural activity, generating speech sounds while also giving real-time audio feedback identical to his voice. Even though we had no clear sample of his true speech, researchers succeeded in not only decoding exact phonemes but also paralinguistic features, including emotions, tone, and even melodies. This groundbreaking discovery can enable such ALS patients to express themselves with their natural speech and emotions.[3]

BCIs also pose certain ethical challenges. These include users' autonomy, privacy, and responsibility. This raises concerns about user control, sensitive brain data, and a lack of accountability. Such ethical implications can be overcome by introducing varying levels of user control.[4] While most research focuses on training BCI systems to interpret brain signals better, a key question remains: can users also learn to control their brain activity to improve BCI performance? For this, in a recent study, 15 healthy people were taught how to use a BCI system that reads EEG signals. Participants had to imagine syllables daily for 5 consecutive days. Over time, they got better at using BCI, but at different paces. A control experiment was also run, and found that giving participants real-time feedback on how well they are doing is very crucial. People only improved when they knew how the system was responding to their brain signals. As they improved, visible changes were seen in their brain waves. Theta activity increased in their frontal, and gamma activity decreased in their temporal part of the brain.[5]

Another overlooked aspect is the underreported user experience. Knowing how the user feels using BCI devices from soon after surgery to long-term use can help develop better systems. Even though recent advancements in BCI suggest a promising future, such devices are still quite costly, which can exclude patients, leading to unfair gaps and widening societal inequality.[6]

To conclude, there is no doubt that ongoing breakthroughs in neurotechnology hold potential for significant impact. But to make real-world change, we need continued research, user-friendly designs, and attention toward ethical challenges. Urgent investment is required to scale these solutions and ensure equitable access to every patient in need, regardless of their background or income.

 

MissingLinkText

Conflict of Interest

None declared.

• References

• 1 Pirasteh A, Shamseini Ghiyasvand M, Pouladian M. EEG-based brain-computer interface methods with the aim of rehabilitating advanced stage ALS patients. Disabil Rehabil Assist Technol 2024; 19 (08) 3183-3193
• 2 Card NS, Wairagkar M, Iacobacci C. et al. An accurate and rapidly calibrating speech neuroprosthesis. N Engl J Med 2024; 391 (07) 609-618
• 3 Wairagkar M, Card NS, Singer-Clark T. et al. An instantaneous voice-synthesis neuroprosthesis. Nature 2025; (e-pub ahead of print)
• 4 van Stuijvenberg OC, Samlal DPS, Vansteensel MJ, Broekman MLD, Jongsma KR. The ethical significance of user-control in AI-driven speech-BCIs: a narrative review. Front Hum Neurosci 2024; 18: 1420334
• 5 Bhadra K, Giraud AL, Marchesotti S. Learning to operate an imagined speech brain-computer interface involves the spatial and frequency tuning of neural activity. Commun Biol 2025; 8 (01) 271
• 6 Khan S, Kallis L, Mee H. et al. Invasive brain-computer interface for communication: a scoping review. Brain Sci 2025; 15 (04) 336

Publication History

Article published online:
09 August 2025