Edgar Cervantes / Android Authority
TL;DR
- We were able to enable video analysis functionality in Google’s Gemini chatbot.
- The feature allows users to ask a variety of questions about an uploaded video.
- Gemini’s analysis seemed pretty accurate in our short time with the feature.
Google has added plenty of features to Gemini since its launch, and we recently discovered evidence that the chatbot will gain the ability to analyze videos. Now, it looks like we managed to enable the feature.
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.
We dug into a recent version of Gemini for Android (version 16.15.38.sa.arm64 beta) and managed to enable video uploads and analysis. Check out the video below to see how the feature works.
We asked Gemini to analyze three separate video clips. The chatbot’s analysis of the first video is generally correct, as it accurately guessed the clip’s location based on building architecture and other geographic clues. Gemini also seemed to analyze the two other clips accurately.
We were also able to glean a few more details about the video analysis feature. You can actually type a question to go with your video upload, for one. Furthermore, we discovered that your individual and combined video uploads top out at five minutes. Upload a longer video, and Gemini will warn you that you’ve exceeded the maximum duration. Check out the screenshots below.
This isn’t the first time we’ve heard about video analysis in Google’s chatbot as we found strings related to this feature last month. We also discovered evidence earlier this month that Google could let you upload and analyze multiple videos at once.
In any event, the fact that we’re able to unofficially enable this feature suggests that an official reveal isn’t far away. It seems like a safe bet that Google could potentially announce video analysis for Gemini at its I/O event next month.
This capability could be handy for a variety of situations. You could theoretically upload Zoom call recordings to get summaries, upload a screen recording to get troubleshooting help, and more.
