What you need to know
- Google’s Gemini app can now handle audio uploads on Android, iOS, and web, a feature users have been asking for the most.
- Supported formats include MP3, M4A, and WAV, with the app transcribing audio, summarizing key points, and extracting actionable insights.
- Users can upload up to 10 audio files at once, but their total length can’t exceed 10 minutes, and other Gemini usage limits still apply.
Last month, signs popped up that Google was working on letting the Gemini app handle audio uploads. This much-requested feature is now live across Android, iOS, and the web.
The update supports MP3, M4A, and WAV files. Once you upload, Gemini will transcribe the audio, pull out the key points, and give you a clear summary (via 9to5Google).
This feature can be accessed via the plus menu on Gemini’s mobile app or “Upload files” on the web. Once you upload an audio clip, the app analyzes it, turning meetings, interviews, lectures, or voice notes into easy-to-digest summaries and key takeaways.
Top user request comes to life
Josh Woodward, VP of Google Labs and Gemini, shared on X that this has been the feature users have asked for the most.
However, according to Google’s support page, you can upload up to 10 audio files at once, but their combined length can’t exceed 10 minutes. Other Gemini usage limits still apply, so keep that in mind before sending a batch of files.
The audio upload limits aren’t infinite but are fairly generous compared to video. Free users get 10 minutes for audio, which is double the five-minute video cap. Meanwhile, paid users get three times the one-hour video limit.
Another limit to keep in mind is the file count. You can upload up to 10 files per prompt, and this covers everything from code folders with up to 5,000 files to GitHub repos and ZIPs with up to 10 compressed files. The new audio feature counts toward this 10-file total, so it doesn’t expand the overall limit.
Beyond transcription, Gemini can highlight key points, distinguish speakers, and pull out action items or quotes. This, in turn, makes any audio file a neatly structured, searchable document.
