AI WORLD HQHow to Translate Menus, Signs and Labels with Your Phone and Gemini
Point your phone camera at any foreign text and read a clear English translation in seconds.
In short: Point your phone camera at any foreign text — a menu, a street sign, a product label — and Gemini will read it and translate it into English in seconds. It usually handles common languages well, though decorative fonts, handwriting, or blurry shots can trip it up.
In a few minutes you'll know how to translate almost any printed text in the world using just your phone and Gemini — Google's AI assistant. This guide is for complete beginners, and assumes you've never opened the app before. All you need is a phone, the app, and a steady hand.
💡 Tip: tap a step’s number when you finish it — a green tick appears and your browser remembers how far you got.
- The Gemini app installed (free from the Google Play Store on Android, or the App Store on iPhone).
- A Google account signed in on your phone — the same one you'd use for Gmail or YouTube.
- Camera permission turned on for the app (your phone will ask the first time).
- A working internet connection — Wi-Fi or mobile data. Some translation features need the cloud to work.
- Roughly 2–3 minutes to translate one thing end-to-end.
A quick honest note: Gemini's camera translation is good for everyday situations, but it won't always be perfect. For anything safety-critical (food allergens, medication labels, legal notices), treat the answer as a starting point and double-check with a human.
Open Gemini and switch to camera mode
Tap the Gemini app icon on your phone — look for a four-pointed spark or star in blue and white. The main chat screen opens with a text box near the bottom. Find the camera icon — it usually sits right next to the text box, or tucked inside an attachment menu (a + or paperclip). Tap it once. A small panel slides up with options like Take photo or Open camera live. If you only see one option, that's the one to use.
You'll know it worked when your phone's viewfinder fills the screen, showing what the back camera sees.
Frame the text and capture it
Hold your phone about 20–30 cm from the menu, sign, or label — close enough that the text fills about two-thirds of the screen. Try to keep the phone flat and steady, and avoid harsh shadows across the words. If you chose live camera, just point and hold; if you chose photo, tap the round shutter button at the bottom of the screen once. Stay still for a moment after the click so the image isn't blurry.
You'll know it worked when Gemini shows the photo (or live video feed) on screen, with a text box ready for your question.
Ask Gemini to translate
Right below the photo, you'll find a text box — type a short, direct instruction. Mention the source language if you know it, and always say the language you want the answer in (English is the default, but you can ask for others). Press the send icon (an arrow or paper plane). Within a few seconds, Gemini replies with the translated text, usually noting what language it spotted.
You'll know it worked when Gemini's reply contains a clear English version of the text, with any extra details you asked about.
Ask follow-up questions
If anything is still unclear — a single word, an abbreviation, a cultural reference — just type a follow-up question in the same chat. Gemini remembers the photo and its first answer, so you can dig deeper without re-snapping the picture. Ask things like "what does 'al dente' mean here?", "is this dish usually spicy?", or "which one is the cheapest soup?" You can also ask it to rewrite the translation over the photo if you want a visual overlay.
You'll know it worked when Gemini responds with a specific clarification, using the same photo as context — no need to upload anything again.
Save or share the result
Once the translation looks right, you can keep it, copy it, or send it to someone else. Look for small icons near Gemini's reply — usually a copy icon (two overlapping squares) and a share icon (an arrow pointing up out of a box). Tap copy to paste the translation into Notes, Messages, or a translation app. Tap share to send it to a friend who's still puzzling over the menu across the table.
You'll know it worked when the translated text appears in your Notes app (or wherever you pasted it), ready to read at your own pace.
- Tilting the phone so the text is skewed. Hold the phone flat and face the text straight on. If the photo is already taken at an angle, ask Gemini to "read the text even though it's skewed" — it usually copes.
- Photographing a dense full page and getting a muddled answer. A single packed menu is hard for any AI to parse cleanly. Photograph one section at a time, or ask Gemini to "list only the items under 'Pasta'".
- Trusting the translation without checking the important bits. Gemini is accurate most of the time, but it can misread a price or miss an allergen. For anything critical (medical, dietary, financial), confirm with a human or an official source.
- Forgetting to mention the source language. If the menu mixes two scripts (say, Japanese and English), tell Gemini which part to translate. Otherwise it may try to translate the bits you don't need.
Grab one thing with foreign text near you — a food packet in the cupboard, a souvenir, or even a foreign-language webpage on another screen. Snap it with Gemini, ask for a translation, and read the answer. Two minutes, done.
❓ Quick questions
How long does this take?
About 6 minutes — the guide has 5 steps, and you can tick each one off as you go.
Which tool do I need?
This guide uses Google Gemini — but the approach works very similarly in other AI assistants.
Do I need to prepare anything?
- The Gemini app installed (free from the Google Play Store on Android, or the App Store on iPhone).
- A Google account signed in on your phone — the same one you'd use for Gmail or YouTube.
- Camera permission turned on for the app (your phone will ask the first time).
- A working internet connection — Wi-Fi or mobile data. Some translation features need the cloud to work.
- Roughly 2–3 minutes to translate one thing end-to-end.
A quick honest note: Gemini's camera translation is good for everyday situations, but it won't always be perfect. For anything safety-critical (food allergens, medication labels, legal notices), treat the answer as a starting point and double-check with a human.
What mistakes should I avoid?
- Tilting the phone so the text is skewed. Hold the phone flat and face the text straight on. If the photo is already taken at an angle, ask Gemini to "read the text even though it's skewed" — it usually copes.
- Photographing a dense full page and getting a muddled answer. A single packed menu is hard for any AI to parse cleanly. Photograph one section at a time, or ask Gemini to "list only the items under 'Pasta'".
- Trusting the translation without checking the important bits. Gemini is accurate most of the time, but it can misread a price or miss an allergen. For anything critical (medical, dietary, financial), confirm with a human or an official source.
- Forgetting to mention the source language. If the menu mixes two scripts (say, Japanese and English), tell Gemini which part to translate. Otherwise it may try to translate the bits you don't need.
Keep reading
✦ Original step-by-step guide by AI World HQ's AI editorial team. Written in plain language, reviewed for accuracy.
← Back to all stories