Real-Time Sign Language Translation More Accessible
by Devin Coldewey – Techcrunch.com – Sign language is used by millions of people, but unlike many languages, there’s no automatic translation available for those who can’t use it. SLAIT claims the first tool for general use, which can translate around 200 words and simple sentences to start — using only a computer and webcam. Emails and text are useful and common now. But they aren’t a replacement for face-to-face communication, and there’s no easy way to turn signs into written or spoken words. SLAIT is a startup built from research done in Germany, where Antonio Domènech built a small ASL recognition engine using MediaPipe (which offers an effective, open-source method for tracking hand/finger positions) and custom neural networks. Evgeny Fomin and William Vicars joined him to start the company. They built a system that could recognize individual ASL gestures and simple sentences. Translation occurs offline, usually in near real time. They plan to make it available for educational and development work, expanding their dataset so they can improve the model before attempting more significant consumer applications. With the amount of video calls going on these days, accessibility is being left behind — only some platforms offer automatic captioning, transcription, summaries, and none recognize sign language. But with SLAIT’s tool people could sign normally and participate in a video call naturally rather than using the neglected chat function. Fomin says, “There is a lot of work to do to reach a comprehensive library of all sign language gestures. We are committed to making this future state a reality. Our mission is to radically improve accessibility for the Deaf and hard-of-hearing communities.” For now the product is a proof of concept, but what a proof it is — with a bit more work SLAIT can provided something that deaf and hearing people have been wanting for decades.