Looktel Recognizer a big! disappointment?
Submitted by Dan on Saturday, February 25, 2012.
iOS & iPadOS
Hi, I may be alone on this one. I am disappointed with this app. My biggest problem is that you have to create an image library first. This deveats the purpose of this type of app wear you should be able to point your i os devie's camera, take a picture or scan the bar code and have the device tell you what the product is. Another thing is the companhy is saying the is fast and accurate and gives independence. Where is the independence if you first need someone sighted to help you create this image library. Of course the app is going to work fast and accurate. All it is doing is accessing a library that all ready exists. As far as using Apple ivs to help with image librtary, the problem is we all have different things. I think the app is being promoted wrong. The app isn't all that because it is using something that has to be created first. If this company wants to make a truly nice and great app look at Digit-eyes app. The Money Reader is a great app, but the Looktel Recognizer when you really think about it isn't all that amazing.
Hello I fully agree with you. The app misses usefulness completely. The app to me is a big disappointment too. I didn't find the app to be that useful in real life situations, The barcode scanner is much faster then Digiteyes though. That was the only saving grace for the app, they dropped the ball on this one as far as usability goes, I won't be using this app in every day life. dissapointedly, Ed
I think you may be missing the angle that Look Tell was going for with the Recognizer. I'm assuming you were expecting it to be like a barcode scanner that uses object recognition instead to find any object whatsoever. So you could, say, go into a grocery store and pick a can off the shelf and immediately tell what it is by pointing the camera at it. Recognizer, however, is marketed more as a solution to personal labeling items for the blind - at least as I can tell. You won't need the huge database of images of every product on the market that users have submitted - you only want the ones you periodically need to recognize or discern between when you are at home. They also make you record your own message to tell what something is - usually something seen in a labeling system. Also, object recognition using images is likely a lot more resource intensive than merely searching for a UPC number in a database, as well. Thus, it'd be impractical to stuff the app full of images when someone using it isn't going to even use half of it or even a quarter of it. Think of how many images it'd have to compare against - it'd lose the snappy recognition that we've come to expect from the Look Tell software. I rather like the fact it's got very fast recognition and in real-time - no having to search a database over wi fi or data plan. The image repository is obviously meant to hold your personal cache of products you need to recognize on a daily basis or that you purchase a lot. Also, Look Tell is a small company focused solely on apps for the blind - not a huge market. They can't afford to keep a database of images even if it was user-maintained like Digit Eyes' database of UPC codes and labeling. Or at least I am assuming that. Yes, you do have to have sighted assistance to start your image repository but it's a personal labeling thing - you need to personalize it. It doesn't take that long to do, either, if you sit down in one session of maybe a half hour to an hour and have a sighted friend or family member capture the unique aspects of products you commonly need to discern. It even works with logos on t-shirts as well as cans and packages. I personally find it a lot more convenient to use than Digit Eyes as I don't have to struggle to find the barcode every time. Hopefully the ability to share images with other users will fill the gap that the lack of a huge database has opened up. That way you can pick and choose images you may actually use to prevent the app from slowing down with too many images to compare against. Anyhow. Long story short - it's version 1! And it's an awesome app, in my opinion, to use for personal labelling of packages, t-shirts, CDs, DVDs and the like - much easier than sticking braille or button labels on everything. The barcode scanner part of it is sort of an after thought addition - it's not meant to be used as an all-around ID Mate kind of scan-everything deal. The software would slow down too much with millions of images to sift through - but it's great as a personal labeler.
I agree with Kristen on this one. I think that you're expecting the app to do things it was not intended to do. Think about it this way. If I have a reader that I have to pay for, or someone in my family who is willing to sit down and go through common household items as well as clothing, I only have to have this person do this task for me one time. Once it's done, it's in my database, and I can then access it whenever I want to. It's just one of many tools that can be used to help gain independence. If you need something recognized right away, and do not have sighted asistance to do the inicial labeling, there are always other apps like omoby, vizwiz, etc. So, I think you may find it disappointing because you were expecting the app to do things it was not designed to do. Granted, the shared library idea may only be fantastic in theory, but it's great for a lot of other things as well.
Based on my limited use of the app so far I'd say this is a great app, not sure what the shared library idea is going to evolve into, but I've been taking images (with some sighted help) of my clothing and this app is going to save me a lot of time with organizing and recognizing garments. I know there are methods for labelling and organizing different articles of clothing, but I find it tedious and time consuming, after I'm done getting everything into my photo library it'll be a breeze to pick out the right stuff. If all I get out of it is that and other ideas I have about work and neighborhood landmarks, I'll consider the ten bucks as very well spent. I know it's more than we pay for a lot of apps but come on, it's still just 10 bucks - less than the price of a pizza, and has a lot of potential for future improvements.
Absolutely, right. I believe there has been a lot of misunderstanding on behalf of people's part in what Looktel original ideawas supposed to be about. They already realized that Viz Wiz and OMoby has that corner of the market when it comes to being to identify items through a method of taking an image. Wait. Hope you get a hit on your results. This process takes time and data for this to work. Looktel wanted to offer a compromise version of these two wonderful products. Get real time results now with the set of your things you use on a everyday basis. Matter of fact I encourage people to use Viz Wiz and Omoby to help you identify the items you may need to identify. From this results I can put this in to my Recognizer. There are huge potential in to this app that other couldn't do. Can those other products tell you how long to cook a particular frozen meal? Can it tell you that this remote is for the family room and not for my bedroom? Can it tell you that my toothbrush or my wife? You may or may not have these issues nor have any desire to be that specific but the possibility is there. In so many different ways you can use this for. It probably isn't for everyone. I can understand that. Everyone might have their own routine but I have two kids and believe me this is saving my life. Lord only knows what my kids put what where. Things never get put back where they supoose to be. Labeling is such a pain in my butt and I really don't want to label my toothbrush! ;) I believe Looktel hit a home run with this app. It is unique and one of a kind that you aren't find anywhere else. Why would you want to compete for things that already exists? besides I think it works great with Viz Wiz and Omoby myself. They literally go hands and and hand.
Just for the record, the Digit-Eyes database is not user-maintained. We had considered the option, but when we looked at crowd-sourced data of this type on other sites, we determined that it wasn't generally very accurate The reason there are a lot of records (26, 024,994 as of midnight last night) is because early in our product's life, we determined that getting deep content related to UPC / EAN codes was of critical importance to our users. There is no authoritative source of this sort of data, so we had to do something a little different. The D-E database is obtained by using spiders (automated programs that read manufacturer's and retailers sites, retrieve data and store it). This is how Google builds their database as well. We do allow users to submit content, but for accuracy, it is post-processed after submission to ensure the description is accurate for the number and to "drive" the spiders to obtain full content for the item.
Hi Nancy, what do you do with user entries for non-english products? Can you check them, too?
Yes, we check entries in all languages. Digit-Eyes is actually designed for multinational use. This is due, at least in part to the fact that the primary development team is composed of a Dane, a Brazilian and an American. The Dane happens to live and work in Brazil, our Brazilian member lives and works in Canada and the American in Texas, so we are familiar with multilingual requirements! But, on a more serious note, thanks to a dedicated cadre of translators, Digit-Eyes is actually offered in 10 languages English Spanish Portuguese German French Norwegian Danish Swedish Polish Italian We collect product descriptions in all those languages as well plus Latvian, Romanian and a couple of others. When someone scans a UPC or EAN and the code is sent to the host to look up, we check the location of their ISP and determine the primary language for the country they are in. When we search for the content for their code, we give precedence to their native language, use English second if the primary language is not English and then whatever we have.
Dear All, After reading all the comments on this subject, I am amazed that most people forget the past, unless the USA was vastly different than for us in the UK... I remember back to the late 1980's when it was either Braille or plastic sew-on shapes to recognise your clothes and a plastic card to identify the difference between a £1 and £5 note! The app's we have now, like the Look-tel products and Digit Eyes are a revelation to me as a blind person of over 23 years. I would rate these apps alongside the screen readers I was using in 1990 like Outspoken and a bit later, an early version of JAWS. For example, I can now recognise the whole of my CD collection of over 2,000 CD's with only about 7 hours work and I didn't even need a sighted supporter, as my CD collection already had Braille labels. Indeed, my wife commented last night; 'look how far we've come in 20 years, from a Braille embossed label to a digital alternative' and she's sighted! Digit Eyes too is a brilliant app. I only purchased it yesterday because before, my impression was that it was too expensive but I only paid £13.99, which is a little more like it. I spent a couple of hours checking it out and was able to recognise about 20 products with no failures. I was using Red Laser, which is okay but doesn't recognise non-standard products very well but Digit Eyes was excelent. I was able to recognise about 7 products from Sainsburys own range and 5 from Tesco's own range and Red Laser couldn't do that. Indeed, Digit Eyes could even recognise the bar codes on the CD's I tried like Genesis Seconds Out, Yes, 90128 and Asia, Omega, so for me, this is a great app. When I get home from work tonight, I'm going to try the print label part of the app to see if that works. In summary then, Don't forget the past, our lives are much easier now, use these app's in combination to get the best results and look at the relatively low cost of these app's compared with buying lots of sew-on labels and miles of Braille dymo-tape! Many thanks, Steve.
Hi everybody, I also suggest you to try the new release of AudioLabels. Here you can find what are the main skills of this app: AudioLabels helps visually impaired persons to recognize objects through an audio description associated to barcodes or QR codes. It works in two simple steps: 1. Point your camera towards a label with a barcode or a QR code. AudioLabels will search many providers in Internet for an available text description. If a description is found, AudioLabels will read it through VoiceOVer. If no description is found, AudioLabels will ask you to enter a text or to record an audio description. 2. When you point again the camera towards that label, AudioLabels will read the description. With AudioLabels you can contribute to extend the database of available barcode descriptions. Whenever you add a new description you can choose whether to share it with other AudioLabels users or not. Shared descriptions will be available for all AudioLabels users. The objects you want to recognize do not have a printed barcode or QR code (like the CD with your files)? Don’t worry: with AudioLabels you can receive your custom labels by email so you can print and stick them on every object you want to recognize. You can use AudioLabels to recognize boxes, DVDs, clothes, paper documents, etc. The app is now sold at a one-month-lasting promotional price of 9.99$. For those who already use AudioLabels the update is free. Here is our website: http://www.everywaretechnologies.com App Store: http://itunes.apple.com/us/app/audiolabels/id496513473 We'd be very pleased to hear something about it, so your suggestions and questions are very welcome.
Dear Audio Labels poster, Yes, I have used audio labels and found it useful. However, it is very difficult to use the bar code labels, which are downloaded from the site because you need a sighted person to help you cut them up, even if you use sticky labels! I wonder if the bar code labels could be made to fit standard sticky labels, something like those that are used for address labels? In the UK we use standard labels for all sorts of things, some very small that could take a single Audio Labels bar code. Is this possible? Kind Regards, Steve.
Hi Steve, at the moment with AudioLabels you can't print labels onto a sheet of self-adhesive labels, but you can only print them on adhesive sheet and then cut them with the help of a sighted person. For example we suggest you to ask for more than one label sheet (PDF) and ask a printing house to have the labels printed and cut (so you can print a lot of labels at once). Although we are taking into account your request and we'll work on it. Thank you