Ryan Haines / Android Authority
TL;DR
- In the latest beta of the Gemini Android app, we spotted an option to attach audio files, such as MP3s, to chats.
- A “Talk live about this” prompt also appears, but the audio isn’t accurately processed yet.
- While the feature might not be fully operational, we know that Gemini can comprehend audio.
Ever wished you could just throw an MP3 at Gemini and have it explain what it’s about? That might be coming soon, as we’ve spotted early signs of audio file support in the Gemini app for Android.
⚠️ An APK teardown helps predict features that may arrive on a service in the future based on work-in-progress code. However, it is possible that such predicted features may not make it to a public release.
In version 16.30.59.sa.arm64 of the Google app beta, we managed to activate a new file attachment feature when chatting with Gemini. You can now attach audio files like MP3s, and when you do, Gemini shows a new suggestion: “Talk live about this.” That sounds promising, but it doesn’t actually work yet.
After uploading an audio file, you can either type a regular question or choose to “talk live” about it. In both cases, Gemini doesn’t seem to understand or respond to the file in any meaningful way. Sometimes it ignores the audio entirely. Other times, it confidently makes things up, as seen in the third screenshot below, but chatbot hallucinations aren’t exclusive to audio files or Gemini.
Still, it’s not hard to see where this is going. On the developer side, Gemini already supports audio input through the API. You can feed it audio and ask it to describe what it hears, summarize it, or transcribe what’s being said. It even handles timestamp requests like “from 2:30 to 3:29,” and works with formats like MP3, WAV, and FLAC.
That’s likely what Google is building toward on the Android app — we’re just not there yet. For now, it’s more like a placeholder than a finished feature, and there’s also no guarantee when or if it will launch. However, with image uploads now widely available in the Gemini app, audio support seems like a logical next step.
Thank you for being part of our community. Read our Comment Policy before posting.
