by Gilles P`epin
From the Editor: Gilles P`epin is President of VisuAide, Inc.
It is a great pleasure for me to be here today and have the opportunity to talk to you about a subject of great interest to me, a subject which has been discussed over and over, but about which there is still much to be said. I'm talking about access to information for blind people.
My presentation today will focus on the question: What is the next step toward better access to information for blind people? So I have entitled my presentation "Access to Information: The Next Step Forward."
The First Steps
One of the landmark technological achievements toward independent access to information, after the advent of Braille, was the introduction of the Talking Book on cassettes. The arrival of the Talking Book meant a major cultural breakthrough for the blind community. Over the years it became for many the key to accessing education, culture, and knowledge, but for quite some time the Talking Book has undergone very few technical improvements.
But, more generally, in the last two decades tremendous progress has been made in the quest for better access to information. First, at the beginning of the '80's access to personal computers offered great opportunities and high expectations. Computerized Braille production also greatly improved access to an ever-growing number of documents.
Then, at the end of the '80's OCR [optical character recognition] technology began a new era in the access-to-information field. As a result blind users have had increased access to the world of print. In this context, at the beginning of the '90's a worldwide struggle began, and the main efforts of this industry have centered on access to Graphical User Interface for the past eight years or so.
Then came the Internet, like a gift from heaven. And access to the Internet has been a great leap forward, with more and more documents now available in digital format.
The Next Step
But what is this next step in our quest for equal access to information? Sure, we all know that Braille production tools, reading systems, personal computer access, and access to the Internet will get better, faster, and cheaper in the next few years. And that is great.
But what else can make more documents accessible? What else can make production of alternate-format books more efficient and universally accessible?
The ideal access to information for blind users should be every bit as convenient, as dependable, as complete, and universal as it is in the sighted community. With the extraordinary technologies at our disposal, nothing less is acceptable.
So what is this next revolution toward access to information? I for one firmly believe the answer to this question is the Digital Talking Book. You will say that I am biased since I have been working developing this technology for the last few years, and you are certainly right--I am biased. Nevertheless, I believe the Digital Talking Book technology will be something very important for blind people in the near future.
Before I go into more detail about the Digital Talking Book, let me tell you a little about our efforts at VisuAide in this battle for equal access to information. For the last eleven years we have been involved in the development of innovative technologies for the blind. Our quest to develop information access led us at the end of the last decade to develop IRIS, one of the first reading systems on the market after the Kurzweil Reading Machine. We also developed Proverbe, a high-quality French software speech synthesizer, and Liber, a forty-cell Braille display.
In the field of digital speech compression, we developed a few years ago a product called Magnum--the first digital recorder for the blind and predecessor of Victor, our digital talking book player, which we'll talk about in a few minutes.
VisuAide is still actively involved in research and development projects, and I think we have an interesting approach to the development of new technologies for the blind. I would like to share it with you.
As you know, the design of new products for the blind community is often based on state-of-the-art technologies. This was certainly the case when refreshable Braille, speech synthesizers, and reading systems appeared. At the same time it is very difficult to justify such developments on the blindness market only. Our industry simply cannot afford such important development.
So at VisuAide we decided to look at the matter from a different angle. Here is how we have decided to do it. When we believe technology can provide solutions to some of the problems faced by the blind population, we simply find other markets such a technology can serve. Then we either create a new company focusing on this technology or find a partner interested in the new market.
So about three years ago we created a sister company known as Haptech Technologies. The focus of Haptech is to add a tactile dimension to computer use, what we call the haptic sense. Remember when Dr. Cranmer described yesterday his experience in a plane? That is the haptic sense.
Just a little anecdote here: when we started the Haptech company, we had an initial brainstorming meeting, and the first idea that was launched was to make Playboy accessible for the blind through the haptic feedback device. We think this technology will be highly useful for expanding information access in the blind community in the future. We are now developing a product called the PenCat and a software program called TouchWindows. In the future these tools will be used to virtually touch graphical objects displayed on a computer screen.
But the broader market that Haptech is focusing on is input/output devices for specialized CAD systems, for controlling music systems settings and for learning tools. If we had undertaken this research within VisuAide, we would never be where we are right now.
More recently, about eight months ago, we created a second spinoff company developing City Go, an electronic tourist guide based on a Global Positioning System (GPS) and a cleverly referenced information database to give tourists pertinent information about the places they are visiting.
I am sure you already know how useful a modified version of this tool can be to blind users. The product will be called Mira and will be a simple, hand-held orientation tool able to guide you to your destination. We believe that, because this development is for the general market, we will be able to make the product better, smaller, and cheaper.
Finally, as Deane Blazie explained without my permission, we are also researching new technologies for the Braille cell of the future.
What Is a Digital Talking Book?
So now lets come back to the main subject of this presentation, the Digital Talking Book or DTB. As George explained earlier, a DTB is simply a set of files which, put together, forms a multimedia document. One file contains the text of the book with all the indexing tags that describe the structure of the book. The indexing is done with the XML mark-up language.
A second file contains the audio that has been recorded the same way we've done with cassettes, except that now the recording is digitized and probably compressed later on in the production process. Finally you have a SMIL file, which contains the necessary data to synchronize the text and the audio.
We could also take some time to discuss in more detail the document type definition, the navigation control center, and the resource file associated with the DTB, but these are not necessary to understand the basic concept.
As George also described, these three basic files may include different levels of contents to form different types of books to answer different user needs. For example, the richest form of book will contain all the text with its entire structure heavily indexed, all the audio recorded in a high-quality mode, and a SMIL file linking every recorded audio clip to the corresponding structure element in the book. If we look more closely at this richest type of DTB, we see that it can be reproduced for the user in all the alternate formats presently in use and more.
First, since the audio is now in Digital format and linked to the structural elements of the book, the DTB is an improved Talking Book. Second, because the DTB includes all the text with the necessary tags for describing its structure, it can be made into a good Braille book using the future Duxbury software to translate it into Grade II Braille and format it properly. The text file of the DTB is also an improved version of the E-Text. And, finally, the same DTB without any changes can be used by low-vision users as well.
As you can see, the DTB is a very versatile book format that has been designed taking into account the needs of all these clienteles. This new concept, the DTB, was made possible because of an extraordinary combined effort by people around the world, including specialized libraries, producers of alternate-format books, manufacturers, and users within the DAISY Consortium and the NISO working group.
These efforts are leading to the establishment of a universal standard. From the manufacturers' point of view it is important that one, and only one, standard be adopted. Then we can relate to something concrete to design the best possible playback tools. I would like to add that our participation over the last three years in the DAISY Consortium and the NISO group has been very gratifying. This close relationship with all the stakeholders of this new technology enabled us to design a product that really meets the needs of all the interested groups involved. I believe that the DAISY and the NISO groups' working model is one we should reproduce in the development of other technologies.
DTB Playback Tools
Now that we know what this DTB is all about, let's take a look at which playback tools the user will have access to. As I described earlier, because of the fact that the indexed text is included in the richest type of DTB, blind users will have access to a large number of tools.
First, for Braille users: the DTB will be distributed on a CD or over the Internet and can be read simply by using Braille translation software on your PC to print in Braille directly or by using your Braille display to access it. You will also be able to use your portable Braille Reader, on which you have downloaded your favorite DTB. The best example of such a portable reader would be your Braille Lite equipped with the appropriate reading software. And we are working with Blazie Engineering on this.
Another interesting product in this category is the BookWorm from HandyTech in Germany, distributed in the U.S. by Arkenstone. It is an eight-cell Braille Reader designed for reading documents only. It has nice navigation and scrolling functions to review the document.
Again, because of the inclusion of the text in the ideal DTB, speech-synthesis technology can also be used. Just like their Braille counterparts, portable DTB speech players will also be available. The Braille 'n Speak can certainly be used for reading a DTB.
There is also a very neat little product called the Road Runner. It is a tiny device that you can hold in the palm of your hand. It's equipped with a good-quality speech synthesizer and two MB of memory to store the text file of the DTB. This is a dedicated speech reader, but it was not developed for the blind although it seems to be quite easy for blind persons to use. In fact, it was designed for professionals who would like to download Internet content and read it on the road.
It's worth noting that large-print users will also be able, when provided with the right software, to read the DTB using the enlargement they need. Portable devices will also be available in this category. In fact, I believe that the Open E-Book readers will offer such functions as enlargement as part of their basic functions. These Open E-Book readers include Rocket E-Book at $499 from Nuevo Media and the SoftBook, $599. I would like you to notice that our Victor is only $495, which makes it the cheapest E-Book player on the market today.
Audio DTB Players
Last but not least is the audio DTB player category. These players will use audio and the synchronization with the text to offer you all the efficiency and comfort you need while reading a book. In this category we find two types of players: software players for use with your computer and hardware players.
There are actually three software players on the market. The first is called Sigtuna. It was developed by a U.S. company, Productivity Works, under a contract with the Japanese Society of Rehabilitation for People with Disabilities. This software is distributed free of charge by the JSRPD.
The second software, PlayBack 2000, was developed by Labyrinten Data, a Swedish company, under a contract with TPB, a Swedish Library for the Blind. It will also be distributed for free. Finally, Productivity Works is also offering its own playback software called LpPlayer.
Although these software programs are free of charge or very cheap, they offer most of the basic functions blind persons would need to read a novel, a magazine, and the like. But there is still a big gap where more powerful software is needed to answer the needs of students and professionals reading heavily indexed documents.
Finally, Arkenstone would like to get involved in this next step toward better access to information. Arkenstone, together with other partners within the DAISY Consortium, is searching for grants to adapt its software to the DTB world to serve the needs of learning-disabled and dyslexic users, still using this same DTB we're discussing.
Now the Audio Hardware Players
There are two Digital Talking Book players on the market. These players are focused on playing back the audio file to the user, just as a tape recorder does. They don't use text files that much, but they offer a wide range of navigation functions as I will be demonstrating in a few minutes.
One of these players is called Plextalk. It is manufactured by Plextor, a well known Japanese company. Plextalk is a desktop model based on a caddie CD-ROM drive. The advantage of the caddie is that the user does not have to handle the CD directly. The caddie is a protective case that you insert directly into the player. My lawyers have warned me that this is all I am entitled to say about this competitor product, but, seriously, you are welcome to see it. I will have it with me for the rest of the day.
But there is another DTB player I can talk about for hours and hours. It is called Victor and is manufactured in Canada by VisuAide. The product is distributed in the United States by Blazie Engineering.
Victor, a Portable DTB Player
It is really a great product. But, before I present it to you in more detail, let me tell you about the design process we followed at VisuAide to develop this DTB player.
In 1996 the European Blind Union began a consultation process with users and libraries to set out the characteristics of the ideal digital Talking Book player. The result was a very useful document entitled "Reaching Forward to the Twenty-first Century: User Requirements for the Next Generation of Talking Books."
This document describes in great detail the needs of blind users. For example, it says that the DTB player should include sixty-five different navigation and search functions that are useful or highly desirable and also states that it should have no more than eight keys. Otherwise senior citizens would be intimidated by it. It also says that batteries should last for at least twelve hours; it must include a CD-ROM drive; and, of course, it should fit in your shirt pocket.
Then the NISO working group was formed. We were really enthusiastic because here was a chance to find the right design criteria. So at the first meeting we began a discussion about batteries. Should they be rechargeable; should they be standard batteries; should they be replaceable by users; should they be solar batteries; and so on. After three hours of discussion someone finally said, "Let the manufacturers decide the battery type." Then we moved to something more important, a snoring detector. Yes, we discussed for an hour the fact that a lot of people read in bed, and the player should stop reading when the user falls asleep.
But seriously, as I said before, our participation in the DAISY Consortium and in the NISO working group was extremely useful in the design of our Victor product. So at the end of these two processes we felt we knew all--and I really mean all--the needs of users. And we knew we had to make some compromises. But which?
So here are the choices we made while designing Victor. And by the way, we used smaller groups of users to help us to make the final choices.
First of all we wanted the machine to be as compact as possible. Since we wanted it to be portable, we had to choose a CD ROM drive as opposed to a caddie drive. We decided that blind people could manipulate CD's because they have been doing that with music CD's for quite a long time. Our product runs on rechargeable batteries for about six to eight hours and weighs less than one kilogram.
Victor is also a flexible digital Talking Book player. The fact is that, when we designed the product, we said that the technology will change quite a bit in the near future, so we needed to have a flexible design because in the future you are going to have new speech-compression algorithms coming out. You are going to have new DAISY, new NISO, and new Open E-Book formats to support. And you are going to have new functions that the users will ask for. So we designed Victor around a very flexible platform. It is based on the RISC processor, called a Strong Arm, and it operates with a Linux operating system. The Victor is also quite user-friendly. To see the features demonstrated, see me afterward.
Now that we know for sure that DTB is the next step toward better access to information, I would like to share with you my thoughts of what the near future will bring. First, we will see improvements on the players side, adding exciting new features like the possibility of adding your own voice notes to your document, more powerful search and navigation functions, the ability to have words spelled or their definitions given as you are reading.
The second generation of DTB players will be hand-held devices without any moving parts. You will be able to connect to the Internet, navigate a catalogue of available Digital Talking Books, and choose the one you like; it will be downloaded over the Internet and stored in the player's memory.
We may also see in a few years mainstream products become very good DTB Players. For example, another project going on in Japan consists of developing DTB playback software that will run on a Windows CE new gaming machine. The MPEG-3 music players may also be good DTB players in the future.
So in conclusion I would like to say that the DTB technology is a giant step forward toward equal access to information. I believe it is in the best interests of blind people to begin producing books in this new format as soon as possible.