Meta’s new AI models can recognize and produce speech for more than 1,000 languages

Meta has built AI models that can recognize and produce speech for more than 1,000 languages—a tenfold increase on what’s currently available. It’s a significant step toward preserving languages that are at risk of disappearing, the company says.

Meta is releasing its models to the public via the code hosting service GitHub. It claims that making them open source will help developers working in different languages to build new speech applications—like messaging services that understand everyone, or virtual-reality systems that can be used in any language.

There are around 7,000 languages in the world, but existing speech recognition models cover only about 100 of them comprehensively. This is because these kinds of models tend to require huge amounts of labeled training data, which is available for only a small number of languages, including English, Spanish, and Chinese.

Meta researchers got around this problem by retraining an existing AI model developed by the company in 2020 that is able to learn speech patterns from audio without requiring large amounts of labeled data, such as transcripts.

They trained it on two new data sets: one that contains audio recordings of the New Testament Bible and its corresponding text taken from the internet in 1,107 languages, and another containing unlabeled New Testament audio recordings in 3,809 languages. The team processed the speech audio and the text data to improve its quality before running an algorithm designed to align audio recordings with accompanying text. They then repeated this process with a second algorithm trained on the newly aligned data. With this method, the researchers were able to teach the algorithm to learn a new language more easily, even without the accompanying text.

“We can use what that model learned to then quickly build speech systems with very, very little data,” says Michael Auli, a research scientist at Meta who worked on the project.

Source