[ad_1]
Millions of people use sign language, but methods of teaching this complex and subtle skill have not developed as quickly as written and spoken languages. SLAIT aims to change school with a computer vision-assisted interactive tutor that allows aspiring ASL speakers to practice at their own pace, like any other language learning app.
SLAIT in 2010 It launched in 2021 as a sign language AI translator (hence the name): a real-time video chat and translation tool that recognizes the most common signs and helps an ASL speaker communicate more easily with someone who doesn’t know the language. But early successes faltered when the team realized they needed more time, money and information than they could afford.
“Initially we got good results, but after several tests, we now realized that there is not enough data to provide full language interpretation,” explained Evgeny Fomin, CEO and founder of Slate. “We had no opportunity for investment, we had no opportunity to find our backers, because we were stuck without a product launch – we were in trouble. Capitalism is… hard.
“But then we thought, what can we do with the technology we got from R&D? We realized that we had to make an educational solution, because our educational technology is definitely enough,” he continued. It’s not that there are minimum standards for people to learn, but that the fluidity and subtlety of fluent and mature sign language are commands that are much harder to capture and interpret than one or two words at a time.
“We found a very talented person (CTO Nikita Nikitin) who helped us develop a product and built SLAIT School. And now we have our first customers and some trailers! “
Current online sign language courses (here’s a solid list if you’re curious) are generally very traditional. You have lectures and demonstrations, vocabulary and examples, and if you pay online, you can have someone video review your work. It’s high quality and mostly free, but it’s not quite the interactive experience people expect from apps like Duolingo.
SLAIT School uses an updated version of gesture recognition technology that powers the translator display app to provide instant feedback on words and phrases. Watch it in video form, then experiment until you get it. It’s currently only available for desktop browsers, but the team is planning a mobile app as well.
“We have some room for improvement, but that’s exactly what we plan to deliver. Students can access the platform, practice, make gestures, interact with an AI tutor, and it costs them about an hour with a physical tutor,” said Fomin. Mobile apps that we intend to make free.
Users can take the first few lessons to see if the system works, then it’s $39 per month, $174 semi-annually, and $228 annually. Compared to large-scale language apps supported by capital and other business models, the price may seem high, but Fomin emphasizes that this is a new and unique category, and real-world educators are their primary competition.
Image Credits: SLAIT SCHOOL
“We will actively communicate with users and try to find an economic model that makes the best prices and subscription plans affordable. We really want to make the platform free, but we haven’t had the opportunity to do that yet. Because this is a great product… we need to make a stable economic model that works as a whole.”
Traction is the flywheel of technology and content for the company. By collecting data (with explicit opt-in permission, which the community is more than happy to provide), they can expand and improve the curriculum, and refine their gesture recognition engine.
“We see two directions for growth,” said Fomin. “The first is to cover more language groups, such as British Sign Language and Japanese Sign Language. We also want to make the curriculum more adaptable or to provide a curriculum for medical and scientific signs. If we have enough investment to grow and scale, a world-leading platform to communicate with a doctor in sign language. We can be.
After that, he said, “Maybe we can finally develop a translator. We can cross this fence!”
[ad_2]
Source link
