The speech-to-text feature is there now sealed in all current computers. But what if you didn’t have to dictate to your computer? What if you could write while thinking?
Silicon Valley startup Sabi is coming out of the closet with this goal in mind. The company is developing a wearable device that decodes a person’s internal speech into words on a computer screen. CEO Rahul Chhabra says its first product, a brain-reading cap, will be available by the end of the year. The company also designs a version of the baseball cap.
This technology is known as brain-computer interface, or BCI, a device that provides a direct communication path between the brain and an external device. While many companies, such as Elon Musk’s Neuralink, are developing surgically implanted BCIs for people with severe mobility disabilities, Sabi’s device could allow anyone to become a cyborg.
This isn’t quite Musk’s vision of the future, which includes implanted brain chips to allow humans to connect with artificial intelligence. But venture capitalist Vinod Khosla, who was an early investor in OpenAI, says a non-invasive wearable device is the only way to convince many people to exploit BCI technology.
“The biggest and worst use of BCI is the ability to talk to a computer by thinking about it,” says Khosla, founder of Khosla Ventures, one of Sabi’s investors. “If a billion people are going to use BCI every day to access their computers, it can’t be invasive.”
Sabi’s brain reading cap relies on EEG, or electroencephalography, which uses metal disks placed on the scalp to record the brain’s electrical activity. Decoding imagined speech from EEG is already possible, but is currently restricted to petite sets of words or commands rather than continuous, natural speech.
Sensors line the inside of the hat and non-invasively read brain signals.Photo: courtesy of Sabi
The disadvantage of wearable systems is that sensors must listen to the brain through layers of skin and bone, which dampens neural signals. Surgically implanted devices receive much stronger signals because they are so close to the neurons. Sabi believes that the way to boost the accuracy of the wearable is to significantly boost the number of sensors in its device. Most EEG devices have from a dozen to several hundred sensors. The Sabi cap will be equipped with between 70,000 and 100,000 miniature sensors.
“Given the high density of detection, it pinpoints exactly what and where neural activity is happening. We use this information to get much more reliable data to decipher what a person is thinking,” Chhabra says.
The company aims for a starting typing speed of around 30 words per minute. That’s slower than most people type, but he says speeds will improve as users spend more time with the limit.