Noisy environments are a challenge for deaf people, especially when trying to understand what a specific person is saying in that environment. A US startup has now created glasses designed to help — providing the user with real-time captions.
A team of investigators led Nirbhay Naranga computer science student at Cornell University, developed glasses for the deaf — which show subtitles of what other people are saying.
The technology will now be commercialized by , a spin-off of the university.
With an appearance similar to that of a pair of traditional glasses With a thick frame, the device is connected via Bluetooth to an application on the user’s iPhone. The company is, however, creating a version of the app for Android.
As the phone microphone picks up voice of a person located close to the user, a speech to text algorithm AI-based converts your spoken words into text that is relayed to the glasses.
This text is projected onto the inside of the lenswhere the user can read it, and it is also displayed on the cell phone screen, as a backup copy.
The application currently recognizes languages such as EnglishSpanish, French, Italian, Chinese, German and Portuguese. Users can manually add technical terms, slang or other words that are not in your vocabulary.
Cornell University
Nirbhay Narang, a 25-year-old IT student, created smart glasses that use AI to provide real-time transcriptions of conversations.
“On passionate about human-computer interaction“, said Narang, a computer science graduate from the university’s Faculty of Arts and Sciences. “I consider myself more of a designer than an engineer and I have friends who have been working in the smart glasses sector for some time.”
According to the statement, one battery charge is enough for five hours of use, and the glasses can also subtitle movie conversations you cinema us.
The glasses are currently available for sale — for now, only in the United States — through the company’s website, where they are listed with a starting price of 699 dollars (around 650 euros).
It is important to highlight that the use of the transcription algorithm speech to text has a additional cost. Buyers can choose a $49 monthly plan that allows them to use the system offline — or pay 2 dollars per hour to use it through an online server.
