During the Inside the Lab: Building for the metaverse with AI livestream event on Wednesday, Meta CEO Mark Zuckerberg gave a glimpse of what metaverse will look like. He shared the latest progress Meta AI is making in different areas like self-supervised learning and computer vision. Zuckerberg revealed that the company aims to build an AI-powered “universal speech translator” for metaverse that works for everyone in the world. Also Read - Facebook revamps groups, introduces chat channels, audio channels and more
As per the company, almost 2 billion people or about 25 percent of the world speak a language that doesn’t have a translation system available. To resolve this issue, Meta is working on a two-fold project. The first part is No Language Left Behind and the second is Universal Speech Translator. Also Read - Meta introduces Instagram Reels APIs for developers: How it will work
As per the company blog, “The first is No Language Left Behind, where we are building a new advanced AI model that can learn from languages with fewer examples to train from, and we will use it to enable expert-quality translations in hundreds of languages, ranging from Asturian to Luganda to Urdu. The second is Universal Speech Translator, where we are designing novel approaches to translating from speech in one language to another in real-time so we can support languages without a standard writing system as well as those that are both written and spoken.” Also Read - WhatsApp's upcoming feature is aimed at making shopping easier for you: Check details
The metaverse should be accessible to everyone, no matter what language you speak 🗣️ we’re teaching AI to translate hundreds of spoken and written languages in real time.
— Meta (@Meta) February 23, 2022
Meta faces data scarcity which is a huge obstacle when it comes to expanding translation tools across the world. “MT systems for text translations typically rely on learning from millions of sentences of annotated data. Because of this, MT systems capable of high-quality translations have been developed for only the handful of languages that dominate the web. Expanding to other languages means finding ways to acquire and use training examples from languages with sparse web presences.”, wrote Facebook AI Research in a company blogpost.
With these projects, Meta aims to improve translations so that people who speak different or even lesser-known languages can socialise in metaverse easily.