Fb dad or mum firm Meta Platforms on Tuesday launched an AI mannequin able to translating and transcribing speech in dozens of languages, a possible building-block for instruments enabling real-time communication throughout language divides.
The corporate mentioned in a weblog submit that its SeamlessM4T mannequin might assist translations between textual content and speech in almost 100 languages, in addition to full speech-to-speech translation for 35 languages, combining expertise that was beforehand out there solely in separate fashions.
CEO Mark Zuckerberg has mentioned he envisions such instruments facilitating interactions between customers from across the globe within the metaverse, the set of interconnected digital worlds on which he’s betting the corporate’s future.
Meta is making the mannequin out there to the general public for non-commercial use, the weblog submit mentioned.
The world’s greatest social media firm has launched a flurry of principally free AI fashions this 12 months, together with a big language mannequin known as Llama that poses a critical problem to proprietary fashions offered by Microsoft-backed OpenAI and Alphabet’s Google.
Zuckerberg says an open AI ecosystem works to Meta’s benefit, as the corporate has extra to achieve by successfully crowd-sourcing the creation of consumer-facing instruments for its social platforms than by charging for entry to the fashions.
Nonetheless, Meta faces related authorized questions as the remainder of the business across the coaching knowledge ingested to create its fashions.
In July, comic Sarah Silverman and two different authors filed copyright infringement lawsuits towards each Meta and OpenAI, accusing the businesses of utilizing their books as coaching knowledge with out permission.
For the SeamlessM4T mannequin, Meta researchers mentioned in a analysis paper that they gathered audio coaching knowledge from 4 million hours of “uncooked audio originating from a publicly out there repository of crawled net knowledge,” with out specifying which repository.
A Meta spokesperson didn’t reply to questions on the provenance of the audio knowledge.
Textual content knowledge got here from datasets created final 12 months that pulled content material from Wikipedia and related web sites, the analysis paper mentioned.
© Thomson Reuters 2023
