Goal to break language barriers with AI

Mark Zuckerberg, the CEO of Meta, the company that controls Facebook, wants to use artificial intelligence to break language barriers around the world (AI). Meta has announced an ambitious AI-powered initiative that will be instrumental in the development of the Metaverse. According to the company, it is developing a universal voice translator, as well as an AI-powered virtual assistant.

“The ability to interact with anyone in any language is a capability people have always dreamed of, and AI will give it to us in our lifetimes,” Zuckerberg said in an online presentation.

Building for the metaverse, according to the company, will require major advances in AI. The first step is to promote inclusion through the use of language and translation.

Meta AI heralds a long-term effort to build language and machine translation capabilities that will cover most of the world’s languages, according to the company. Two more projects have been added to the mix. “We will use it to enable expert-quality translations in hundreds of languages, from Asturian to Luganda to Urdu,” the corporation stated, “and we will use it to enable expert-quality translations in hundreds of languages, from Asturian to Luganda to Urdu.

The second project is the Universal Speech Translator, in which Meta is developing unique methods to translate speech from one language to another in real time, allowing it to cater for languages ​​that do not have a regular writing system, as well as those that are written . and spoken.

For those who speak English, Mandarin, or Spanish, today’s web apps and tools may seem like they already have the translation technology we need. Today, approximately half of the world’s population cannot access online content in their native language. A unique system capable of translating into all written languages ​​is known as No Language Left Behind. In a blog post, the company stated: “We are also working on Universal Speech Translator, an artificial intelligence system that enables instant speech-to-speech translation in all languages, even those that are widely spoken.”

See also  macOS Sonoma Features: Best for Productivity, Creativity, and More

Machine translation (MT) systems are rapidly improving, but they still rely significantly on learning huge volumes of textual data, therefore they do not work well for low-resource languages, such as those without training data or lacking a system. standardized writing. , depending on the business.

Project CAIRaoke is a novel form of artificial intelligence that powers chatbots and personal assistants. It is possible that one day people will be able to have more seamless interactions with their virtual assistants thanks to this technology.

“Five years ago, we could translate into a dozen languages,” Zuckerberg said at the time. Three years ago we were in 30 languages ​​and now we are targeting hundreds this year.”

Despite being an ambitious project, Meta did not provide timelines or milestones, making it that much more confusing.

Also, using AI to break language barriers is not a new concept. Tech companies like Google and Apple have been working on this, and part of this includes translation tools. India has been working on a similar effort called the National Language Translation Mission, as has the Indian government (NLTM).

The goal of NLTM is to develop automatic speech-to-speech translation and a Unified Language Interface (ULI) for Indian language translation, bringing together multiple projects in the field of Indian language technologies.

Subscribe to our latest newsletter

To read our exclusive content, sign up now. $5/month, $50/year

Categories: Technology
Source: vtt.edu.vn

Leave a Comment