According to @AIatMeta, Meta has introduced the Omnilingual Automatic Speech Recognition (ASR) suite, delivering ASR capabilities for over 1,600 languages, including 500 low-coverage languages previously unsupported by any ASR system (source: https://go.meta.me/f56b6e). This breakthrough expands AI-driven transcription and translation services to underserved populations, paving the way for inclusive digital communication, real-time voice interfaces, and new global business opportunities in sectors like education, customer service, and accessibility. With comprehensive language coverage, the Meta ASR suite positions itself as a foundation for next-generation AI applications targeting emerging markets and diverse linguistic communities.

According to Greg Brockman (@gdb), OpenAI Codex is now generally available, offering a new SDK and advanced enterprise features designed to accelerate AI-powered software development. The announcement highlighted a live demonstration where Codex was used for 'vibe coding' of a controller for stage lights, showcasing its integration with a voice interface. This advancement enables businesses to rapidly build and deploy AI-driven applications using natural language commands, unlocking practical efficiencies in automation, software customization, and voice-controlled systems (source: Greg Brockman on Twitter, Oct 6, 2025).