Ryan Haines / Android Authority
Tl; DR
- In the latest beta version of the Gemini Android application, we spotted an option to attach audio files, such as MP3, cats.
- A prompt “speaking live” also appears, but the audio is not yet treated with precision.
- Although the functionality is not fully operational, we know that gemini can include audio.
Have you ever wanted to be able to throw a MP3 on Gemini and have it explained what it is? It could happen soon, because we spotted the first signs of supporting audio files in the Gemini application for Android.
⚠️ A APK decay Help predict the features that can happen on a service in the future depending on the current labor code. However, these predicted features may not be public release.
In version 16.30.59.SA.RM64 of the Google App Beta, we managed to activate a new feature of file attachments during the cat with Gemini. You can now attach audio files like MP3s, and when you do, Gemini shows a new suggestion: “Talk live on this subject”. It seems promising, but it doesn’t work yet.
After downloading an audio file, you can either type a regular question or choose to “speak live”. In both cases, Gemini do not seem to understand or respond to the file significantly. Sometimes he completely ignores the audio. Other times, that invents things with confidence, as we can see in the third screenshot below, but the Hallucinations of the Chatbot are not exclusive to audio files or Gemini.
However, it is not difficult to see where it is going. On the developer side, Gemini already supports audio entry via the API. You can feed him audio and ask him to describe what he hears, summarize it or transcribe what is said. He even manages horoditing requests like “from 2:30 p.m. to 3:29 am” and works with formats like MP3, WAV and FLAC.
This is probably what Google builds on the Android application – we are just not there yet. For the moment, it looks more like a reserved space than a finished feature, and there is no guarantee either when or if it is launched. However, with image downloads now widely available in the Gemini application, audio support seems to be a next logical step.
Please be part of our community. Read our comment policy before publishing.
