Seeing AI is a free app available on both the Google Play Store for Android devices and the Apple App Store for iPhone and iPad. It uses your device’s camera and an internet connection (Wi-Fi or mobile data) to process visual information using artificial intelligence.
It is highly versatile and works well with screen readers and magnification, whether these are turned on or off. This means the app can fit around the accessibility settings you already use on your smartphone or tablet.
If you are not a screen reader user, look out for the Play button. This reads aloud any text that the app generates, which can be especially helpful for reading longer pieces of text if you are tired or experiencing eye strain.
Please note: the layout of buttons and controls is different on Apple and Android devices.
The app received a major update in 2025 and is now more accurate than ever. Although Seeing AI has been available for several years, its performance and features have improved significantly. As with all AI tools, however, it is important to remember that results are not always 100% accurate.
How to pronounce “Seeing AI”
Seeing AI is written with no spaces.
Phonetic spelling:
Sierra – Echo – Echo – Indigo – November – Gold – Alpha – Indigo
Getting Started
After installing the app, the first time you open each feature (called a “channel” or “mode”), Seeing AI will provide spoken and on‑screen instructions on how to use it. You can access this help at any time by tapping the question mark button.
The app will ask for permission to use your camera and microphone on some devices. It is important to allow this, as the app will not function correctly without access to these.
App Layout Overview
Following the 2025 update, the app is now organised into three main modes:
- Read
- Describe
- More
When you open the app, it always starts in Read mode.
Read Mode
Read mode is designed for reading text quickly and automatically.
As soon as you open the app, it begins looking for text in front of the camera and starts reading it aloud. This can feel surprising at first, but it is very useful when speed matters – for example, when trying to locate and read a sign in a public place.
For more accurate reading of documents such as letters or forms, it is recommended to turn document detection on in the settings. Hold your phone or tablet above the page and follow the spoken cues to line up the document before taking a photo. This method reduces errors caused by hand movement and gives much clearer results.
Describe Mode
Describe mode allows you to take photos and receive detailed descriptions of what is in the image.
You can:
- Take a selfie to hear a description of yourself and your background
- Use the rear camera to describe rooms, landscapes or objects
- Import photos from your camera roll (for example, pictures shared in family WhatsApp groups)
A newer feature allows you to ask follow‑up questions about an image to get more detail.
It is worth noting that the AI is not always accurate when identifying a person’s age or gender.
You can also train this mode to recognise important people, such as family members, so the app can tell you who appears in photos.
Example 1 - Nature Scene
AI description:
The image depicts a natural scene with a small waterway surrounded by dense vegetation.
- Foreground: The waterway occupies the central portion of the image, with tall reeds and grasses growing along its edges. These plants are densely packed and extend vertically, some leaning slightly into the water. The water reflects the surrounding greenery and light from above.
- Middle ground: On both sides of the waterway, there are clusters of bushes and shrubs. These are interspersed with trees that have visible trunks and branches. The trees are positioned close to each other, creating a canopy of leaves that partially obscures the sky.
- Background: Beyond the waterway and vegetation, there is an open view of a larger body of water, possibly a lake or river. The water appears calm and stretches horizontally across the image. A narrow strip of land, possibly a beach or shoreline, is visible between the vegetation and the larger body of water.
- Spatial relationships: The trees and bushes frame the waterway, creating a natural corridor that leads the eye toward the larger body of water in the background. The vegetation is dense and closely packed, with the waterway cutting through the centre.
Example 2 - Coastal Scene
AI description:
The image depicts a coastal scene with three distinct areas. In the foreground, there is a patch of green grass with scattered vegetation and small plants. The grass transitions into a pebble‑covered beach that occupies the middle section of the image. The pebbles are interspersed with sparse patches of grass and low‑lying vegetation. Beyond the beach, the sea extends into the background, with the horizon line separating the water from the sky. The sea appears calm, and the sky above is overcast with clouds.
More mode – additional tools
The More mode contains a range of extra tools that many people find helpful:
Colour
- Describes colours of objects. Works best with bold colours and is less reliable with pale shades or complex patterns. Moving the camera slowly can help identify individual colours within patterns.
World
- A feature still in development. It can detect items such as doors in real time, includes a proximity sensor, and can detect nearby people. It is designed to support, not replace, mobility aids such as a white cane or guide dog.
Product
- Scans barcodes to identify items such as food or toiletries.
Currency
- Identifies banknotes and tells you their value. Currently supports 17 different currencies, making it especially useful when travelling.
Find My Things
- Allows you to train the app to recognise specific items, such as your keys, and guide you to them if they are misplaced. Setting this up can take some time and may be easier with help from a sighted friend or family member, particularly when understanding distances and angles during training.
Person
- Scans your surroundings to detect people, estimate how far away they are, and whether they are facing you. This can be useful when preparing to take photos or navigating social situations.
How people use Seeing AI
Members of the vision‑impaired community have shared many real‑world uses for the app, including:
- Reading hotel room door numbers
- Reading menus in restaurants, including wall‑mounted menus
- Describing landscapes during walks
- Identifying US dollars while travelling
- Describing photos of grandchildren in family messages
- Reading post and letters
Tips for best results
- Place documents flat on a table and hold your phone above them, keeping it parallel to the surface
- Turn on document detection for letters and magazines
- If positioning is difficult, try standing while taking the image
- Enable Manage lighting in settings so the flashlight turns on automatically in low light
- Choose a voice that you find comfortable to listen to
- Adjust the speech speed to suit you
- Remember the help button (question mark) is available in every mode
