Meta为1 600+种语言,包括濒危语言,发布了开放源代码的AI语音对文本工具,可免费用于研究和商业使用。 Meta releases open-source AI speech-to-text tool for 1,600+ languages, including endangered ones, with free access for research and commercial use.

Meta推出了一个开放源码语音到文字AI系统,支持1 600多种语言,包括许多低资源语言和濒危语言,有可能扩大到5 400多种语言。 Meta has launched an open-source speech-to-text AI system supporting over 1,600 languages, including many low-resource and endangered ones, with potential for expansion to over 5,400 languages. 该模型利用基于背景的学习,使新语言能够以极少的例子添加,并通过GitHub和Hugging Face免费提供供研究和商业使用。 The model uses context-based learning, enabling new languages to be added with minimal examples, and is freely available for research and commercial use via GitHub and Hugging Face. 它旨在改善无障碍环境,支持多种口音和区域差异,目的是加强全球通信工具,促进语言包容性。 Designed to improve accessibility, it supports diverse accents and regional variations, aiming to enhance global communication tools and promote linguistic inclusivity.