Facebook is working on a universal translation system that can understand 128 different languages

You know those futuristic translation devices they have in sci-fi movies, that allow characters from completely different universes to talk in real time, without the need for prior knowledge of the local language?

That could soon become a reality, as Facebook’s AI research team today unveiled its latest work on an improved language translation system, which can currently translate up to 128 different languages ​​into English within a single app.

Called XLS-R, the process is able to perform speech recognition, speech translation and language identification at a higher rate than any similar available system.

As Facebook AI explains:

Trained in more than 436,000 hours of publicly available speech recordings, the XLS-R is based on wav2vec 2.0, our approach to self-directed learning of speech representations. Using speech data from a variety of sources, ranging from parliamentary sessions to audio books, we have expanded to 128 different languages, covering almost two and a half times more languages ​​than our predecessor. ”

Indeed, Facebook says it has been tested against XLS-R four major multilingual speech recognition benchmarks, in which it outperformed their scores in most of the languages ​​tested.

“Specifically, we tested it in five BABEL languages, 10 CommonVoice languages, eight MLS languages ​​and 14 VoxPopuli languages.

As you can see here, error rates for these systems are still relatively high for some languages, but XLS-R shows a significant improvement in accuracy, which Facebook continues to improve as it develops the process.

This could eventually open up a range of new applications, including cross-border connectivity and commerce, expanding business opportunities around the world.

Facebook also notes that such systems could be of significant value in its future metaverse applications, enabling a more universal connection in these open-world digital communities.

As noted, Facebook is still developing the system, and Facebook also notes that more than 7,000 languages ​​are spoken worldwide, so it is far from a truly universal translator. But it’s another step toward that next phase, and toward making Facebook a more critical utility in the relationship.

You can read more about Facebook’s XLS-R survey here.

Source link

Naveen Kumar

Friendly communicator. Music maven. Explorer. Pop culture trailblazer. Social media practitioner.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button