Microsoft just came out with a new app for iOS products: the Seeing AI application. This app debuted in July and already has created quite the buzz among people who are visually impaired or blind. The app was designed in part by software engineer Saqib Shaikh who is blind. Because of his impairment, Saqib knew the challenges people face every day with blindness. The app is great because it uses the camera on an iPhone or iPad to describe the world. There are several channels to use in just one app. The best part of the app is that it is completely FREE!
Each channel offers its own advantages. The channels and their description are below.
Short Text Channel: When using the camera to view text such as an envelope or any print, this channel starts reading the text in real time right away. This is a fast way of identifying print. You must keep your hand steady, but it is really fast.
Document Channel: Again you will use the camera to view text in a document and then use VoiceOver to read the text. One part that I really like about this channel is that if you are holding your device over the document but it can’t see the whole document, the app tells you which part it cannot see (such as “left edge not visible”) so you can reposition the camera to get the whole document. It will tell you to hold steady while it takes a photo of the document and then the print will come on the screen. You must use VoiceOver in order to have the document read to you, and you have to have Wifi to process the document. You can also share the document or upload it to your device.
Product Channel: This channel can be used to scan a barcode on items, and it will identify the item. All you have to do is hold the item up to the camera and the device will start beeping when it recognizes a barcode. The beeping gets faster as the barcode comes into view. It will process the barcode and tell you what the product is. It works on all kinds of items, not just food and sometimes will even provide more information about the product such as cooking directions for food items.
Person Channel: This is an interesting part of the app; when pointing the camera at a person, the app will tell you approximately how far away a person is and if he is in the middle of the shot. If you tap the device to take a photo, it will process the photo and tell you the approximate age of the person, the gender, and what kind of expression is on his face. You can then save the photo or share it. I had three women in my office, and we were playing around with the app when it described a woman as being a 34 year old blonde woman (her age was 31 and she had red hair). Another aspect of this channel is that you can upload photos or take 3 pictures of a person and add his/her name in a database so it will recognize him/her in the future. I have a friend who is blind who loves to take photos, and I can see her loving this app because it will tell her if she has the people in the view of the camera before taking the photo.
Scene Channel (Beta Version): The Scene Channel is definitely a version that needs some work. You can use this to take a photo of what is in front of you and it tries to describe the scene. It is not perfect technology yet but will get better with time.
Improvements that are coming: