Originally Posted by
bocastephen
So I've found ChatGPT invaluable (along with some great teachers on Instagram) in helping me expand my Japanese casual conversation skills - but reading is still a problem. My solution was to take photos of signs, menus, etc and upload the photos to get a translation which works great, but is cumbersome.
This weekend I heard that Google Gemini rolled out a new feature for IOS where you could use a live camera view to get a description of what's around you, including translations just by aiming your camera - but I can't seem to find this feature in the Gemini app I installed. It will only prompt me to upload a photo - I already do with this ChatGPT who I've trained on our Japanese interactions, so this feature is useless to me in Gemini.
Has anyone successfully tried to enable Gemini with a live camera view where the camera is open in Gemini and you can just let it look at things and describe them, locate you, or translate text?
Yes, I use Gemini's models as my primary AI (Claude being secondary). The iOS app is probably what's limiting you. Unfortunately, Apple's walled kindergarten severely limits what app developers do (not to mention that Gemini is currently a conflict of interest for their OpenAI partnership). I've always found that Google translate's camera feature worked great for me...live or with a snapshot. Even before AI. I have to imagine Google is leveraging their AI products these days to enhance Translate.