Recently, Google unveiled its 1,000 Languages Initiative, aiming to get to the point of supporting the 1,000 most spoken languages.
Google has published a new article where it is introducing Google USM: Scaling Automatic Speech Recognition Beyond 100 Languages. This is a new framework for multilingual automatic speech recognition (ASR) that could be extended to hundreds of languages.
In this article, we discuss how it works and why important.
A few months ago we talked about Meta’s NLLB, a model capable of translating into more than two hundred languages. Today Google presents USM, but why so much interest?
