The Beta iteration of the Microsoft Edge browser has integrated an AI-powered real-time audio translation capability, enabling the instantaneous translation of videos embedded in web pages. This innovative feature harnesses a local AI model to transform the original audio of the video into the desired target language, subsequently synthesizing speech for seamless output. Presently, it accommodates input in Spanish, Korean, and English, while offering output options in a variety of languages, including Simplified Chinese.

Users can access this functionality by navigating to Settings → Accessibility → Real-time Translation. However, it's important to note that this feature necessitates a minimum of 12GB of RAM and a quad-core CPU for optimal performance. Currently, the translation function operates reliably solely on YouTube. Moreover, during the translation process, the Edge browser will persistently consume a substantial amount of memory.