Google Gemini AI now supports 10-minute audio transcription and task extraction

Josh Woodward, Google’s VP of Gemini, said audio uploads were a top user request, and testing showed high transcription accuracy across formats, with Gemini also extracting key points, to-do lists, and actionable insights from recordings

Google has rolled out a significant update to its Gemini AI assistant, introducing the ability to upload and process pre-recorded audio files for transcription, summarization, and task identification.

The new feature allows users to upload audio clips of up to 10 minutes in length—such as meetings, interviews, lectures, or voice notes—through Gemini’s web or mobile interface. Once processed, the content is converted into searchable, summarized documents directly within the Gemini platform. This tool is separate from Gemini Live, which handles real-time voice interactions, and is instead tailored for users who want to analyze recorded content.

Josh Woodward, Google’s VP of Gemini, stated that audio uploads were among the most requested capabilities. Early testing demonstrated high transcription accuracy across various audio types, from casual phone calls to scripted content, though minor inaccuracies in name recognition were noted. Beyond simple transcription, Gemini can extract key points, generate to-do lists, and identify actionable items from the audio.

Smarter processing for work, study, and everyday use

Expanding on its AI functionality, Gemini now offers more than just text conversion. Users can isolate statements by individual speakers, request simplified summaries, or turn recordings into question sets and study guides. This makes the tool especially useful for students, professionals, and content creators aiming to repurpose recorded material into usable formats.

Despite these upgrades, the service has some constraints. Audio uploads are capped at 10 minutes, and users on the free tier face daily limits on usage. While Google has yet to announce pricing for larger-scale access, the new functionality draws from a user’s existing Gemini quota, meaning careful usage may be required for frequent users.

With this update, Google continues to position Gemini as a practical AI assistant for daily productivity—offering rich audio analysis features that complement its expanding ecosystem of tools and integrations.