My favourite app for visually impaired people

There are a number great apps out there designed specifically to help visually impaired people, but let me tell you why Seeing AI is my absolute favourite. It’s multi-functional, quick and reasonably accurate. One of the reasons I rate this app is everything is in one place. There’s no switching between apps to do different tasks. The app has an impressive 9 different functions, called channels, which all perform a specific task. You can read different sorts of texts and handwriting, get descriptions of people or places, identify colours, currency and products and detect light, all within the same app. Let me tell you how each channel works for me and simply helps me get stuff done.


1.   Short text.Hold the camera on your phone over any printed short text and Seeing AI will read it to you. This is the channel I use most frequently. I use it to read signs, envelopes, use-by dates and other information on food packaging (great for supermarket shopping), price tags, clothing labels, medicine boxes and countless other things. It’s easy to focus and it starts reading right away where other apps might take a while to process before reading.

2.   Document. The app takes a photo of the whole page and then reads it aloud. It gives verbal directions when trying focus on the document like ‘top edge not in view’ so you can adjust the camera accordingly. I find the easiest way to get the position of the page right is to place the camera of your phone on the centre of the page you want to read and slowly move it away, keeping the phone parallel to the page. (Remember the camera is not in the middle of your phone). Eventually the camera will find all the edges and take the photo without you having to fiddle around too much. It’s like scanning a document on your phone. As the app reads the document it cleverly recognises formatting too, so it’ll tell you if it’s about to read a heading or a title. I have used this for reading letters, recipes and articles in print in magazines and newspapers. 

3.   Product. For this to work you must hold the camera of your phone over the barcode of the product you need to identify. Move the box, jar, packet, bottle etc around until you hear a beeping noise. The beeps become faster as you get nearer the barcode to help you locate it. This is pretty useful although it doesn’t always recognise the product. I did have a theory that it wasn’t great at supermarket own brands, but recently it’s failed to recognise some fairly well known brands for me too. The other issue is that if you buy multi packs of products, the barcodes don’t appear on all the individual products so if you buy a multipack of baked beans for example, once you have opened the outermost packaging and put the individual tins in the cupboard, Seeing AI can’t help you identify those tins at a later date. I have some useful vision and I use the product channel to distinguish between products which are in similar packaging like shampoo and conditioner, baby foods, jars, tins – you get the idea. This is a  great function but I wish it recognised more products.

4.   Person. You take a picture of a person or people and the app gives you some details about them, usually how many people there are, an estimated age, gender, what they are doing and even an emotion. For example, “a 50 year old man, sitting on a chair, looking happy”. You can also type the name of the person in after you’ve taken a photo and next time the app sees them will say their name instead of ‘a man’ or ‘a woman’. It’s quite clever really, but I haven’t really found a use for this channel. I feel strange taking pictures of someone without their knowledge, but then it’s equally strange asking someone I don’t know if I can take their picture… maybe it’s just me. Until I find a use for this channel, it’s fun to see how old and how grumpy it thinks people are!

5.   Currency. Hover your camera over a note and it will tell you what it is. Simple and brilliant. The app currently recognises British Pounds, US Dollars, Euros, Canadian Dollies and Indian INR. Indian INR is a relatively new development so here’s hoping Microsoft will continue to add more currencies for our friends who live in other countries and for our own international adventures. This channel doesn’t recognise coins, but I find coins are much easier to identify by touch than notes but if you get stuck, the short text channel will read the words on the coin to you.

6.   Scene. This channel describes a scene you capture with your camera. I get the feeling this technology is in its early days as the descriptions are almost always preceded by ‘probably’. It’s still pretty good but not always as accurate as other channels. I guess the more we use it the better it will get. Take a picture of a scene anywhere and get an idea of your surroundings. Examples Include ‘probably a living room with furniture and a large window’ and ‘probably a girl standing in front of a door’. This is great for getting your bearings and even though it’s not perfect, it’s still impressive!

7.   Colour. Simply point your camera at the item you want to know the colour of and it will tell you. This is useful when shopping for or locating clothes and pairing socks! The array of colours Seeing AI knows is not extensive but it’s enough to give you a good idea. It’s basically like asking my husband what colour a dress is. He’d say brown whereas my best friend might say it’s a warm beige with pinky undertones. 

8.   Handwriting. Works on the same principle as the other text related channels and is wonderful when you receive handwritten cards and notes. I don’t have any other apps that can do this, so I appreciate this one.

9.   Light. I don’t personally use this as I have some useful vision but for those who have no light perception this would be great for determining if a light is on or off without having to touch potentially hot bulbs.

10.                 Bonus tip! Seeing AI can describe photos in other apps! It’s not really a channel but a useful feature. If you see a photo in Twitter or someone emails you an invitation as a picture, SeeIng AI will describe these to you and read any text it finds in the picture. Open the picture in WhatsApp, messages or wherever it is, click on ‘share’. Don’t worry you’re not going to share it, but under the ‘share’ menu there’s other options like print, copy and ‘recognise with Seeing AI’. If it’s not there click on ‘more’ and select it from there so it appears on the menu.

Et voila! There is a lot you can do with this app and it’s completely free! If you don’t already use it, I’d recommend giving it a try. The only snag with Seeing AI is that it is only available on iOS as far as I know. The last update I saw, Microsoft were working on an android version, but I haven’t seen a release date yet. Let me know if you’ve heard any more on this.

This is a good example of how artificial intelligence can improve the lives of visually impaired and blind people. The colour of a dress or the contents of a tin seems tiny in the grand scheme of things, but to a visually impaired person to be able to quickly identify these things on their own can be kind of empowering. It’s great for simple identification tasks but for tasks that require more analysis or the human touch, apps like ‘Be My Eyes’ would be more appropriate. I find that many things that slow me down day to day, are to do with figuring out what stuff is and all I need is quick answer. For this purpose, Seeing AI is a great little tool to have in your pocket.

What’s your favourite app that helps visually impaired people? Have you tried any of the paid apps? Are they worth it? Let me know, I’m really interested to find out what’s out there that I haven’t tried yet.

One thought on “My favourite app for visually impaired people

Add yours

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at

Up ↑

%d bloggers like this: