Google has shared more information about the Universal Speech Model (USM), a system the company describes as a “critical first step” in achieving its goals, which is now moving closer to its goal of building an AI language model that 1000 different languages to beat ChatGPT.
Last November, the company announced its plans to create a language model that will support 1,000 of the world’s most spoken languages, while also unveiling its USM model.
The tech giant describes USM as a family of advanced speech models with 2 billion parameters trained on 12 million hours of speech and 28 billion sentences of text, spanning more than 300 languages.
USM, which is for use in YouTube (e.g. for subtitles), can perform automatic speech recognition (ASR) not only for widely spoken languages such as English and Mandarin, but also under-resourced languages such as Amharic, Cebuano, Assamese, and Azerbaijani just to name a few,” Google said in a blog post.
Google currently claims that USM supports over 100 languages and will serve as the “base” for a much larger system.
Meanwhile, Google is expected to introduce a host of AI features to its products in the near future, including Gboard for Android to integrate the Imagen text-to-image generator.
Read all the latest technical news here
(This story has not been edited by News18 staff and was published from a syndicated news agency feed)