Meta Builds First Speech-to-Speech Translator AI Mainly for Spoken Languages (WATCH VIDEO)

Alt: "Mark Zuckerberg"

Meta AI team has built first speech-to-speech language translator that primarily works for spoken languages rather than written.

Mark Zuckerberg, Founder and CEO Meta Inc., disclosed this on Wednesday on his verified Facebook page.

Zuckerberg who posted a video illustration of how the AI speech translation system works using Hokkien (an only spoken language), said the AI will help native speakers of languages that can not be written verbally communicate with people who do not understand the language. 

Mark made the video illustration with Penguin, one of the researchers working on the project.

The Meta boss said the company is "open sourcing" the system so "people can use it for more languages.

He said: "Meta AI built the first speech translator that works for languages that are primarily spoken rather than written. We're open sourcing this so people can use it for more languages."

With this, Meta becomes the first company to develop system that can translate languages that can not be written.

Google translate can only translate written languages from letters to letter, letter to voice, or voice to letter.


Post a Comment

Previous Post Next Post