The independent identification of U.S. currency by people who are blind or have low vision was not often the center of conversation until a federal judge ruled, in December 2006, that U.S. currency is inaccessible to people who are visually impaired. The resulting media attention and discussion within the blindness community have touched on many aspects of the subject, including using technology as a strategy to identify money. At AFB TECH, our mission is to evaluate technology and to provide information to the community. We recently tested money-identification technology. This article reports our findings, including how well the technology works and whether people who are blind or have low vision find it useful.
What should a portable money identifier be able to do? Besides being able to handle a variety of old and new currency designs and bills in various conditions, a money identifier must work efficiently and accurately; that is, it should not be difficult to use, it should work quickly, and it should not require numerous attempts before it works. No money identifier is worth having if it misidentifies bills. Of course, it should also be affordable and small enough to carry around in your pocket. Is there such a device?
As of January 1, 2007, we were aware of only one device that is intended specifically for personal use to identify U.S. paper money—the Note Teller 2, manufactured by BRYTECH of Canada. Another device, Voice-It-All-USA, which combines money identification, color identification, and a digital voice recorder, was withdrawn from the market pending the update of new currency, according to the web site for the reseller of the device. Alternatives for identifying paper money include the Kurzweil 1000 and OpenBook computer-based scanning/reading packages, which were evaluated in the July 2006 issue of AccessWorld, both of which include a money-identification mode, but we did not evaluate them for this article because they are not portable. Also, as we were finishing this article, we learned that the Kurzweil-National Federation of the Blind Reader, which was evaluated in the November 2006 issue of AccessWorld, has a new software upgrade that includes a money-identification mode. This article focuses exclusively on the Note Teller 2.
The Note Teller 2 is a black, rectangular box measuring 6 inches long by 3 inches wide by roughly 1 inch thick and weighing 8 ounces with a 9-volt battery installed. The top of the device has a slot about two-thirds of the way toward its back edge, which is where you insert a bill to be identified; molded edges along the side of the unit help guide the bill into the slot. The front edge of the unit has a headphone jack for privacy and a slider switch that is used to adjust the volume and to select announcements in English or Spanish. When the unit is silent, the switch selects low, medium, or high volume, and while the unit is speaking, the switch is used to change from English to Spanish, or vice versa. The back edge of the unit is slightly thicker than the front and has a ring, so you can attach the Note Teller 2 to a neck or wrist strap or secure it to a desk or counter. The device comes with a vinyl sleeve for protection while not in use.
Caption: The Note Teller 2.
The Note Teller 2 is designed to recognize all U.S. bills, from $1 to $100, and can be upgraded by the manufacturer to identify new bill designs that may enter circulation. Sliding the end of a bill into the slot activates the Note Teller. It does not matter which end of the bill you slide in or which face is up, but the instructions advise you to insert a bill "swiftly" for the best results. Within two to four seconds, a male voice clearly announces the bill's denomination. If the Note Teller 2 cannot determine the bill's denomination, it says "cannot read." The unit also announces "replace battery" when the battery is weak. If you hear it continually say "out of order," then it is time to send the unit in for repair.
How We Tested
We divided our examination of the Note Teller 2 into two parts: a lab test and a user study. In our lab, three AFB TECH staffers used the device to identify a sample of U.S. bills to measure the accuracy of the Note Teller 2 (see Lab Tests at the end of this article). The lab test also assessed the general ease of use of the device, along with the included documentation. For the user study, we assembled two groups of six people who are blind or have low vision, one group in Huntington, West Virginia, and the other in Dallas, Texas (see The User Study: People Try the Device at the end of this article). We observed the way they used the device and noted their reactions to it.
Documentation, Ease of Use, and Convenience
The Note Teller 2 is one of the easiest devices to learn and use that we have ever tested at AFB TECH. It works right out of the box, and there are no complicated setup procedures. You simply slide a bill into the slot and wait to see what the device says. The documentation is also of the highest quality, is straightforward, and provides an excellent description that was written with the reader who is blind in mind. Although braille is not available, the manual comes in large print and on an audio CD in both English and Spanish. It provides hints for using the Note Teller 2, and we found them to be useful. In particular, a specific note about keeping bills as flat as possible and directions on how to insert bills swiftly were important. Additional information about dealing with creased bills would have made the instructions even more useful.
Although we were impressed with the simplicity and accessibility of the Note Teller 2, our observations revealed some design limitations related to its size. That is, the unit is too large to carry in a pocket, and some people may find it awkward and cumbersome to use in many settings, such as at a checkout counter.
Does It Identify Money Well?
To answer this question, we repeatedly used the Note Teller 2 in the lab using expert testers who operated the device as consistently as possible and in two user studies in which participants who are blind or have low vision encountered the device for the first time.
The device is simple, accessible, and easy to figure out. Despite these pluses, the Note Teller 2 did not overcome its middling money-identifying ability when it was tested a total of 564 times by our experienced evaluators in the lab. It successfully identified bills in four out of five trials, which is, in our experience, on par with using vending equipment that accepts paper money. The condition of bills was the strongest predictor of success, with bills in good condition identified easily and worn or wrinkled bills identified with more difficulty. The Note Teller 2 is confused by $2 bills. Each of the three testers heard the clear and unmistakable voice announce "$50" when they inserted at least one of the $2 bills that were used to test the device. Since $2 bills are rarely found in circulation, their misidentification is not as serious a problem as the misidentification of more common denominations. Nevertheless, this shortcoming could conceivably be costly if you are not aware of it.
While the lab tests presented a picture of a device that compares favorably with other technologies that accept cash, the experiences of the two groups of users told a much more nuanced story. Most of the participants in the Huntington group used the Note Teller 2 successfully most of the time, but several became frustrated while using it, in two instances going so far as to give up trying to identify a bill. When considering the Note Teller 2 as a piece of technology, the group rated it highly. Yet when asked, only two of the six Huntington participants stated that identifying money is both a significant and frequent problem for them.
The experience of the Dallas group painted a somewhat more complex picture. Getting the device to work was much more difficult for these participants. As in the lab, the Note Teller 2 misidentified two bills, but this time they were $10 bills, which the machine declared to be "$100." Despite considerable frustration and the two misidentifications, five of the six Dallas participants said that identifying money is important and rated the Note Teller 2 highly.
All the members of the Dallas group with low vision and two of the three members of the Huntington group with low vision found the Note Teller 2 more appealing than did those who were blind who used nonvisual techniques for everyday activities. One could speculate that they experienced problems with fluctuating vision and certain environmental lighting conditions and that they did not have nonvisual techniques to accommodate the identification of money.
The Bottom Line
The Note Teller 2 is a simple device that worked reasonably well and reliably in the lab when tested by experts, with the exception of identifying $2 bills. But when users tried out the device, it was apparent that it should work much better and more reliably. The two groups of users had what might be termed a mildly positive experience with the device, but did not show the level of enthusiasm that might have been expected had the identification of money been more of a problem for them and had the device worked better. The Note Teller 2 is accessible, but it clearly has limitations in doing what it is designed to do. There is room for improvement in reliability, ease of use for bills in poor condition, convenience, and accuracy. The most disturbing finding was that on two occasions in the user study, the device misidentified $10 bills as $100 bills.
The efficiency of use was much less consistent among members of the Dallas group. Both the number of times the participants gave up on a bill and the number of times a bill had to be reinserted before it was read were much higher than for the participants in the Huntington group. These issues notwithstanding, the Dallas group rated the device highly.
Opinions varied widely among the 12 members of both groups who tried the Note Teller 2. Most rated it highly and expressed serious to moderate concern about identifying money. However, some preferred the techniques they use to identify money over the Note Teller 2 and stated that they are not concerned about identifying money. Like other studies that AFB TECH has conducted, this one included some still unanswered questions. What is the significance of the misidentification of $2 bills and a few $10 bills? If all the participants knew a variety of techniques for managing money, would their opinions and the results have been different? And most important, will many people who are blind or have low vision invest $270 on another assistive technology device that has limitations and is not confidence inspiring in its ability to do what it is supposed to do? Don't bet money on it!
Funding for this product evaluation was provided by the Teubert Foundation, Huntington, West Virginia. We would like to acknowledge the research assistance provided by Lee Huffman, Priscilla Rogers, and the Huntington and Dallas volunteers.
View the Product Features as a graphic
View the Product Features as text
View the Product Ratings as a graphic
View the Product Ratings as text
Product Information
Products: Note Teller 2.
Manufacturer: BRYTECH, 600 Peter Morand Crescent, Suite 240, Ottawa ON K1G 5Z3 Canada; phone: 613-731-5800; e-mail: <inquiries@brytech.com> web site: <www.brytech.com>.
Price: $270.
Lab Tests
To learn about bills that are currently in use, we consulted a local bank. We were told that the distribution of money to banks has changed in the past few years. Rather than packages of new, pristine paper currency from the Federal Reserve, packages of money now contain bills of varying quality and may contain one of several designs that include color and other attributes. Therefore, a teller may give a customer five $20 bills ranging from new to used and that have either the newest colorful designs or earlier designs.
For the lab tests, three experienced users of assistive technology who are AFB TECH staffers (two are blind and one has low vision) used the device to identify a sample of 47 U.S. bills of all denominations, from $1 to $100, including $2 bills. The sample of bills included currency in good condition (crisp and unwrinkled), in poor condition (well-used bills that were wrinkled), and in average condition (between good and poor). Of the 47 bills that were tested, 10 were in good condition, 14 were in poor condition, and 23 were in average condition.
The sample of bills also represented what we determined were the three main categories of currency design: "colorful bills," which include color in addition to the traditional black and green associated with U.S. currency; "off-center-portrait bills," which are characterized by an offset portrait and large numbers on the back; and "center-portrait bills," on which the portrait is centered and that lack large numbers. The "colorful" and "off-center-portrait" bills occur only in $10, $20, and $50 denominations. Twenty of the sample bills were in these denominations and represent all three design styles. Trials with these 20 bills were used to determine differences in success rates among the three styles.
The Note Teller 2 was placed on a table with the bill slot facing the tester, as suggested in the instruction manual. Each of the three testers inserted each of the 47 bills into the device four times: once on each face of the bill and once at each end of the bill. In this way, the Note Teller 2 was tested in a total of 564 trials. Each tester inserted each bill as consistently as possible according to the instructions. Observations were made for each of the 564 trials to determine if the Note Teller 2 identified a bill successfully, misidentified a bill, or said "cannot read."
With the exception of $2 bills, the Note Teller 2 correctly identified all the bills, but not always on each of the four tries of a particular bill. Success rates for the denominations other than the $2 bills ranged from 88% to 93% except for the $10 bills, which were identified 68% of the time. The condition of the bills mattered. Disregarding the $2 bills, overall success rates were 97% for good bills, 82% for average bills, and 74% for poor bills. The testers observed that bills in poor condition required special attention to make sure the edges were as even as possible and not folded. Unlike the condition of the bills, the design of the bills did not affect the success rate. The success rates of the three designs were nearly identical, with the center-portrait bills scoring 79%, the off-center portrait bills scoring 80%, and the colorful bills scoring 81%.
The $2 bills, all of average condition, were particularly troublesome. The Note Teller 2 successfully identified the $2 bills on only 6 of 36 trials for a success rate of 16.7%. Even worse, the $2 bills were misidentified as $50 bills five times, with each bill misidentified at least once. None of the other bills, regardless of its condition, was misidentified during the lab tests.
Return to article, or use your browser's "back" button.
The User Study: People Try the Device
We recruited two samples of six participants, each of which had three persons who were blind and used nonvisual techniques to read and three persons with low vision who used either vision or a magnification device to read. The Huntington, West Virginia, group ranged in age from 25 to 66, and the Dallas group ranged in age from 61 to 100. None of the participants had any experience with the Note Teller 2.
We asked each participant to take part in a series of four trials using the Note Teller 2 to identify a set of five bills in each trial. At the beginning of each session, the participants were asked to listen to the audio instructions on the CD that came with the device. We used bills of the most common denominations that are in circulation ($1, $5, $10, and $20) and used an equal number of bills in good and poor condition. The same bills were used in both testing locations. In addition to recording the number of times it took for the Note Teller 2 to identify each bill successfully, we observed the participants' use of the device and asked the participants several follow-up questions.
Results from the Huntington Users
Although no bills were misidentified, there were three instances in which it took four tries to get the device to recognize a bill and two instances in which it took five tries. Two participants were so frustrated trying to get the device to work that they gave up trying to identify one of the bills after four attempts.
It took four of the six participants (two who are blind and two with low vision) from 30 to 37 attempts to identify the 20 bills, but those attempts included the two instances in which the participants never succeeded in identifying the bills. When asked to rate how the Note Teller 2 compared to their current method of identifying money, two rated the Note Teller much worse than the methods they use, and two rated it much better. The remaining two participants took fewer tries and were successful with each bill. One, with low vision, took 23 tries, and the other, who was blind, took 21 tries. The blind participant rated the Note Teller 2 as equal to the usual method of identifying money, and the participant with low vision rated it as somewhat better.
The two participants who rated the Note Teller 2 the highest had low vision and stated that handling money was a significant and frequent problem for them. When asked how important it was to have a device to identify paper money, the participant who rated this device as very unimportant was blind and expressed complete satisfaction with the current method of identifying money.
Techniques mentioned to identify money included asking a teller or cashier, requesting return change in specific denominations, using vision, holding money for later identification, and organizing money either by location in a purse or by folding methods once bills had been identified. Two of the three participants with low vision stated that they look for the large numbers on newer bills.
Answers to a question about how frequently they experience difficulty handling money varied widely from "not very often, perhaps twice a year" to "all the time." The participants considered the price of the Note Teller 2 ($270) to be high, and several observed that using it in a public situation would be time consuming.
Results from the Dallas Users
On two bill-reading attempts, once for each of two users, a $10 bill was misidentified as $100. One of the bills in question was in good condition with a "colorful" design, and the other was in poor condition with an "offset-portrait" design. In addition, the number of attempts required to identify bills was high for several participants, and these participants exhibited frustration. In one trial, seven attempts were made before the participant gave up. Two other participants gave up after five attempts. The condition of the bills played a major part in effecting success. The bills in poor condition took more than twice as many trials to be recognized as those in good condition.
Four of the six participants (three who were blind and three with low vision) took from 36 to 39 attempts to identify the 20 bills, but these attempts included 8 in which the participants did not succeed in identifying the bills. When asked to rate how the Note Teller 2 compared to the current method they use to identify money, three rated it as much better than the current method, and the fourth expressed a neutral rating.
One participant with low vision had a particularly rough time, taking 51 attempts with three give-ups, but one participant who was blind took only 25 attempts and identified all 20 bills. Both participants rated the Note Teller 2 as much better than the current method they use to identify money.
Four of the six participants rated the Note Teller 2 as excellent or very good, and the remaining two were neutral. Five of the six participants said it was important to have a money-identification device, and one with low vision expressed a neutral opinion. When asked to compare the Note Teller 2 to the method they generally use to identify money, five of the six participants rated it as much better. The participant who provided a different rating was blind and was neutral about the device. The money-identification techniques mentioned by the Dallas participants were similar to those mentioned by the Huntington participants.
The participants stated that they experienced difficulty identifying money from 20 percent of the time to all the time. The grocery store was the most commonly mentioned place where they had difficulty, with half the group mentioning it.
Return to article, or use your browser's "back" button.
