Mind-Reading Beanie Translates Brain Signals
▼ Summary
– Sabi, a Silicon Valley startup, is developing a noninvasive brain-computer interface (BCI) wearable, like a beanie or baseball cap, to decode internal speech into text.
– Unlike surgically implanted BCIs for medical use, Sabi’s device aims to make the technology accessible for everyday computer interaction by a broad audience.
– The device uses high-density EEG with tens of thousands of sensors to read brain signals through the scalp, aiming to overcome the signal weakness of noninvasive methods.
– A key goal is to achieve an initial typing speed of roughly 30 words per minute by interpreting imagined speech, with accuracy expected to improve with user adaptation.
– Investors like Vinod Khosla argue that a noninvasive approach is essential for widespread BCI adoption, as most people will not accept surgically implanted devices.
The ability to control a computer with your mind is moving from science fiction toward reality. A new startup called Sabi is developing a noninvasive brain-computer interface (BCI) designed to translate a person’s internal thoughts directly into text on a screen. The company plans to release its first product, a brain-reading beanie, before the end of this year, with a baseball cap version also in development.
This technology offers a stark contrast to the approach of companies like Neuralink, which focuses on surgically implanted chips. Sabi’s wearable aims to make BCI technology accessible to a broad audience, not just those with severe medical needs. Prominent investor Vinod Khosla, whose firm Khosla Ventures backs Sabi, argues that a noninvasive path is the only way to achieve mass adoption. “If you’re going to have a billion people use BCI for access to their computers every day, it can’t be invasive,” he states. He believes the most significant application is enabling people to communicate with computers through thought alone.
The device functions using electroencephalography (EEG), a method that records electrical activity from the scalp. The core challenge for any wearable EEG system is signal quality, as the skull and skin muffle the brain’s neural signals. Implanted devices get a much clearer signal by being placed directly next to neurons. Sabi’s strategy to overcome this hurdle involves an unprecedented density of sensors. While typical EEG headsets use dozens or hundreds of sensors, Sabi’s cap will integrate between 70,000 and 100,000 miniature sensors.
CEO Rahul Chhabra explains that this high-density sensor array allows the system to precisely pinpoint neural activity. “We use that information to get much more reliable data to decode what a person is thinking,” he says. Current EEG-based thought decoding is limited to small vocabularies or simple commands. Sabi’s ambitious sensor architecture is intended to push toward decoding continuous, natural internal speech.
The company’s initial goal is a typing speed of roughly 30 words per minute through thought alone. While this is slower than average keyboard typing, Chhabra anticipates that speed will improve as users train with the system over time. This development represents a major step toward a future where thought becomes a direct and seamless input for our digital world.
(Source: Wired)
