Hearing with the skin and sending messages with the mind because machines understand the body: For many people, this sounds like little more than science fiction. But for Regina Dugan, such challenges are the science fact of her daily research work.

An engineer with a doctorate, Regina Dugan was the first woman to head up the Defense Advanced Research Projects Agency (DARPA) at the U.S. Pentagon, whose assignments included organizing the first races for autonomous vehicles in the Nevada desert.

After leaving government service in 2012, Dugan took the reins of an Advanced Technology and Projects (ATAP) group at Google. Then, Facebook headhunted her in 2016 to build a new blue-sky lab—the secrecy-shrouded Building 8—for the social network.

Dugan’s team of 60 scientists from myriad disciplines is researching new brain-machine interfaces (BMI). The group intends to create a silent dictation system that captures words directly from the human brain and then sends them as text messages so there’s no need to even reach for your smartphone. At 100 words per minute, it promises to be five times faster than any nimble-fingered typist.

“This is about decoding the words you’ve already decided to share by sending them to the speech center of your brain,” said Dugan, describing the concept at Facebook’s latest developer conference. “As a silent speech interface, it will have the speed and flexibility of a human voice but the privacy of typed text.”

Thanks to the underlying AI technology, the system should function irrespective of your mother tongue. After all, the electrical currents in the human brain transmit not only the elements of language but also the meaning behind the words. “Understanding semantics means that one day you may be able to share your thoughts independently of language. English, Spanish or Mandarin—they all become the same,” says Dugan, outlining her vision.

Equally audacious, her second big idea involves using nerve endings in people’s skin to act as sound receivers just like the cochlea does in the ear. Software would feel language directly and transmit it to the brain without the need to actually hear the words. In this way, the entire human body will become a computer for processing silent communication that breaks through all language barriers.

Silent communication that breaks through all language barriers.