Muscle-reading wearables let you use voice input without speaking a word

As phones and wearable devices become smaller, entering text without a physical keyboard remains difficult. Subvocal offers a hands-free alternative: glasses that turn silent mouth movements into written words. They are intended for people who naturally mouth words while typing or reading, and could make voice-style input possible without speaking aloud. The approach avoids both talking to a device and awkwardly tapping or gesturing at an invisible keyboard.

Instead of using a camera to watch the face, the glasses direct radio signals toward the head and detect changes caused by moving mouth muscles, in a method compared to radar. The technology builds on MIT Media Lab’s AlterEgo, which used electrodes to sense muscle impulses; Subvocal replaces that contact-based sensing with radio. The result is less a conventional pair of smart glasses, with a display or audio features, than a wearable input device that works like a keyboard. Public demonstrations show the system translating silent articulation into text, and it is described as working well.