Startup Aims to Turn Silent Thoughts into Text with a Wearable Device

A Silicon Valley startup, Sabi, is developing a brain wearable designed to convert a person's internal speech into text on a computer screen. The company's first product, described as a brain-reading beanie, is scheduled for release by the end of the year, with a baseball cap version also in development.

The technology uses electroencephalography (EEG) with sensors placed on the scalp to record brain activity. This approach places it in the category of a noninvasive brain-computer interface (BCI).

How the Technology Works

Unlike surgically implanted BCIs being developed by other companies for people with severe motor disabilities, Sabi's device sits on the head. It will contain between 70,000 and 100,000 miniature sensors—a number that vastly exceeds the dozen to few hundred sensors found in typical EEG devices.

CEO Rahul Chhabra stated that this high-density sensing allows for more precise neural activity localization and more reliable data for decoding thoughts. However, wearable EEG systems face the challenge of signal dampening due to the skin and bone layers between the sensors and the brain's neurons. Currently, EEG-based imagined speech decoding is limited to small word sets or commands rather than continuous natural speech.

The Path to Market

The initial typing speed for the device is targeted at approximately 30 words per minute. Venture capitalist Vinod Khosla, an investor in Sabi, believes noninvasive wearable devices represent the only path to widespread BCI adoption.

Khosla described thought-based computer communication as "the biggest and baddest application of BCI."

The company's planned beanie and baseball cap are intended as early steps toward making this form of direct brain-to-device communication accessible to a broad consumer market.