MIT Researchers Work to Capture Silent Speech with Partially 3D Printed AlterEgo Device

Not all of us have the opportunity to have our inner monologue recorded on Twitter or transcribed by a 1950s style secretary, notepad in hand, pen at the ready. The proliferation of recording devices has made it easier to toss off notes on the fly and laptops and phones provide further methods for making our ideas concrete. But what if you want something that will take your thoughts and type them up without all the hassle of having to speak them out loud or hang out with a computer at a Starbucks? Researchers at MIT Media Lab just might have the answer as they have developed a device called AlterEgo that can take those internal thoughts and without the wearer having to vocalize them, read them from your internal verbalizations and jot them down for you.

It’s not just a one-way street with this creation either. The device can also communicate back to its wearer through a specialized headphone that conducts sound directly through the bones of the face and inner ear. In this way, it doesn’t block out signals coming in from ordinary external sources because it isn’t using the same auditory mechanisms, allowing the wearer to hear a conversation and communication from the device at the same time. It it sounds futuristic, that’s because it is; the hope is that the future it creates is one in which the wearer can be more present, as Professor of Media Arts and Sciences, Patti Maes, who has developed the device with her student Arnav Kapur, explained:

“We basically can’t live without our cellphones, our digital devices. But at the moment, the use of those devices is very disruptive. If I want look something up that’s relevant to a conversation I’m having, I have to find my phone and type in the passcode and open an app and type in some search keyword, and the whole thing requires that I completely shift attention from my environment and the people that I’m with to the phone itself. So, my students and I have for a very long time been experimenting with new form factors and new types of experiences that enable people to still benefit from all the wonderful knowledge and services that these devices give us, but do it in a way that lets them remain in the present.”

The device is designed to pick up our subvocalization, the things we say to ourselves, through the placement of a neural network that picks up electrical information from seven locations along the face. They then created a tool that wraps around the neck and has several branches that also touch the face in order to pick up the necessary information. The band was created in photopolymer resin with a 3D printer and uses an internal support system created from a brass rod. In this way, the device remains flexible enough to be comfortable, but still sufficiently structured for the electrodes to retain their position both relative to each other and on the face of the user during regular movement.

When testing AlterEgo, wearers performed some basic computational tasks such as playing a game of chess — during the game they would dictate their moves using the standard chess numbering system and a computer would execute those moves, without the wearer having spoken them aloud. Currently the device has a 92% transcription accuracy rate. That’s pretty impressive given that the words don’t have to be spoken.

The research team is continuing to work on the AlterEgo project and has already developed a second generation model that requires only four electrodes to pick up the subvocalization and Kapur estimates it has a higher accuracy rating than that which they found during their initial usability study. While Kapur and his fellow researchers are remaining modest about their accomplishments, some have already begun to take flight in the the realm of possibilities presented by such a device. Professor in Georgia Tech’s College of Computing Thad Starner theorized that such a device might be extremely valuable to those trying to communicate silently, such as during special ops, or in the midst of extremely loud environments such as air traffic control. Some of those in these kinds of environments, such as firefighters or fighter pilots, are already wearing masks and so integrating a subvocalization headset isn’t such a stretch, as Starner mused:

“A lot of time, special-ops folks have hand gestures, but you can’t always see those. Wouldn’t it be great to have silent-speech for communication between these folks? [Or] for people with disabilities where they can’t vocalize normally. For example, Roger Ebert did not have the ability to speak anymore because he lost his jaw to cancer. Could he do this sort of silent speech and then have a synthesizer that would speak the words?”

While the device isn’t quite there yet, the potential for a multitude of situations is quickly becoming clear and the research team has no intention of resting on their laurels as they continue to push the frontiers of their new technology.

The research behind the development of the device and its creation are outlined in a paper entitled AlterEgo: A Personalized Wearable Silent Speech Interface that was presented at the Association for Computing Machinery’s Intelligent User Interface Conference held in Tokyo.

What do you think of this news? Let us know your thoughts; join the discussion of this and other 3D printing topics at 3DPrintBoard.com.