Typing has always been a bottleneck between what we think and how quickly we can communicate it to a computer. AI-powered dictation tools like Wispr Flow have made things considerably better. You can practically ramble your way through an entire email, or even an article, and have it transcribed into text in seconds.
But there’s a rather awkward problem with talking to your computer. Try doing it in a quiet office, a library, or even your favorite cafe, and more often than not, you tend to attract some attention or end up disturbing others. A startup called The Subvocal Company thinks it has found a solution to this issue. It has developed a pair of smart glasses that promise to turn silent mouth movements into text at over 200 words per minute.
How does it work?
According to the company, its glasses send radar signals through your face. When you silently form words, tiny movements in your speech muscles change how those signals bounce back. An AI model then interprets those changes and turns them into text. There’s no camera involved, so the glasses aren’t trying to read your lips. You still have to deliberately mouth the words, though.
Subvocal says you can use the glasses to write messages, take notes, and interact with AI assistants. It even claims two people wearing them could have private conversations across a crowded room, without anyone else hearing a word.
When can you get your hands on them?
The startup says it has raised $1.3 million from Y Combinator and Afore Capital, and it’s already accepting reservations in the US. The glasses have an early-bird price of $499, with a refundable $199 deposit required to reserve a pair. You’ll pay the remaining $300 later. Prescription lenses will be available for an additional fee, and the company expects shipping to begin in the first quarter of 2027.
Of course, the glasses are still in development, and those impressive typing speeds haven’t been independently verified. I’m also curious how well they would handle different accents, unfamiliar words, or long stretches of silent dictation. Still, if Subvocal can actually deliver what it’s promising, I can see myself using something like this.