US Edition
Scientists Develop Interface to ‘Read’ Thoughts from Speech-Impaired Patients with up to 74% Accuracy
Stanford researchers developed a brain-computer interface decoding inner speech with up to 74% accuracy using a vocabulary of 125,000 words to aid communication for people with paralysis.
- On August 14, 2025, scientists published findings in Cell showing a brain-computer interface decoded inner speech from motor cortex signals with up to 74% accuracy.
- The study involved four participants with severe paralysis who had microelectrodes implanted and were asked to attempt or imagine speaking words to record brain activity.
- Researchers trained AI models on inner speech data allowing the system to decode imagined sentences from a large vocabulary, and included a password mechanism unlocking decoding with a high level of accuracy.
- Lead author Erin Kunz explained that this study marks a breakthrough in identifying the brain patterns involved when a person merely imagines speaking, while co-author Meschede-Krasa highlighted that relying on thought alone to communicate could make the process simpler and quicker for users.
- The findings offer hope for restoring communication in people with speech impairments, though experts caution privacy concerns and note decoding free-form inner speech remains challenging with existing technology.
36 Articles
36 Articles
Stanford research opens the door for people with speech disabilities to communicate only by thinking what they mean.
New Brain-Computer Interface Decodes Inner Speech for Paralyzed Individuals
Via Flickr Researchers at Stanford University developed a Brain-computer interface (BCI) which converts neural signals into spoken words. This BCI, detailed in a study published in Cell, uses sensors implanted in the motor cortex to detect brain activity linked to inner speech. Machine-learning models interpret these signals to predict intended words in real time. These advancements now provide more options for individuals with severe paralysis,…
Science Fiction? Think Again. Scientists Are Learning How to Decode Inner Thoughts
A brain-computer interface has gotten better than ever before at translating thoughts from people with speech difficulties. Researchers are also thinking through how to protect users’ privacy
Brain implants that decode a person's inner voice may threaten privacy
A new brain-computer interface can decode a person's inner monologue. That could help paralyzed people communicate, but also suggests scientists are one step closer to reading a person's thoughts.
Using a special implant, US scientists were able to hear the inner conversation of subjects – if the previously thought "chitty chitty bang bang". However, the error rate is still high.
Stanford's brain-computer interface turns inner speech into spoken words
Four people living with severe paralysis, caused by conditions including amyotrophic lateral sclerosis and brainstem stroke, took part in the trial. One participant could only respond by moving his eyes – up for yes and side-to-side for no. In the study, described this week in Cell, doctors implanted microscopic electrode...Read Entire Article
Coverage Details
Bias Distribution
- 57% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
