Google Gemini rolls out audio uploads for transcription & summarisation

Google has rolled out audio uploads for its Gemini app across Android, iOS, and the web, a feature long requested by users. The tool supports MP3, M4A, and WAV files, and can transcribe recordings, identify key points, distinguish speakers, and generate structured summaries.

Josh Woodward, VP of Google Labs and Gemini, said on X that this was the feature users have asked for the most.

✅ Papercut fixed: You can now upload any file to @GeminiApp. Including the #1 request: audio files are now supported! pic.twitter.com/4Te3xwLC6W

— Josh Woodward (@joshwoodward) September 8, 2025

The upload option is available through the 'plus' menu on mobile or the 'Upload files' button on the web. According to Google’s support page, users can upload up to 10 audio files at once, with a combined limit of 10 minutes for free accounts. Paid subscribers can extend the audio upload length to three hours.

The audio feature follows the AI's broader file handling limits. Most common file types are supported, with up to 10 files allowed per prompt. Audio uploads are capped at 10 minutes for free users, while Pro and Ultra subscribers can extend this to three hours. Video files can be up to 2 GB each, with a total length of 5 minutes for free users and up to one hour for paid plans. Other file formats, excluding video, can be as large as 100 MB per file. Gemini also supports code uploads, allowing one folder or a GitHub repository with up to 5,000 files and a maximum size of 100 MB, as well as ZIP archives containing up to 10 compressed files.

With this addition, the AI platform expands its utility for meetings, interviews, and lectures by making audio content both searchable and easier to digest. Additionally, this rollout makes it one of the few AI apps to support direct audio uploads for transcription and summarisation.