There are several pieces of software available for this purpose, JAWS for Windows, and Window Eyes. There are also speech synthesizers, like Narrator, Flextalk, Access 32, and Eloquence, which turn electronic text into spoken word which comes out of the computer's speakers. There is a product called Kurzweil 1000 which converts text in a book into electronic text, then reads the book aloud by Flextalk software.
A self Contained Talking Browser Especially For The Visually Impaired, Home Page Reader for Windows is a browser for blind and visually impaired users that reads aloud the information on Web pages. It also makes it easy to navigate the Internet, and has many other features, including integrated E–Mail. The browser will automatically switch languages as it encounters them on the page; i.e., pages presented in Spanish are read in Spanish; etc. Home Page Reader uses IBM's ViaVoice OutLoud text–to–speech technology to speak the information clearly. It even recognizes HTML tags and can translate text, graphic descriptions, text in column format and data fields, giving blind people the same access enjoyed by sighted users.
In visiting the page where the grades for the class 78.11 (Intro To Telecommunications) are displayed, after the page is loaded on the screen, I have to wait a moment for an announcement to read on the computer speakers that the location is titling the screen, and Internet Explorer is the application. Then, in JAWS for Windows, I hit the minus at the top right of the PC keyboard. The computer says, "JAWS cursor." Then, to get an idea of what the screen looks like, I combine INS-Down
Arrow, or the Say-All command, and the computer reads out the menus at the top of the screen, the scroll bars, the icons for back, stop, refresh, history, and so on (all boring information, right?), then the title of that page about the grades. It then reads the titles from left to right over the columns of the grade sheet. Next it reads the next
line down, with its Social Security four digit code, and so far as I have experienced, won't scroll down to what is not currently presented on the screen. If I want to hear a cell of a column, I just arrow left and right along that line, landing along the rows across each character of the row. As the cursor passes over each letter or character, it is read out to my ears by the computer. This means I must spell each cell. I have found no way to read columns by word. One must either hear entire columns or spell characters line by line. There is, fortunately, a "say line" feature. But using the grade sheet ends up being quite inconvenient by this navigation method. This is especially so if you don't happen to have a Social Security code that appears on the first screen. Since it won't scroll down, it then becomes necessary to print the document and hope your numbers appear on the printout.
Perhaps later versions of JAWS for Windows will incorporate some way of reading down columns word by word. I hope that the current version of JAWS for Windows, version 3.7, which I have not yet tested as it is so new, does have a way of reading accross columns word by word.
Navigation is done with tab, arrow keys, left mouse button, and without a screen reader, it is impossible for a blind person to use the Internet without sighted assistance. If you have operating system speech Microsoft proprietary products are probably the only ones you will be able to use without much work. Only when you obtain a screen reader, (supplementary third party software which provides additional navigation capabilities and outputs to the blind person what is on the screen) can you go out into the third party applications software.
A blind person tabs through presented text or listens to the whole while sighted person would be looking for context or most relevant paragraphs or links. A sighted person scans the whole Web page with his eyes, and picks out the pertinent links and information, and will follow up on that, going with the flow, perhaps to the original objective that he had in mind, or get sidetracked by finding something he really wants to do, altering plans and gathering other information as presented.
You experienced browsers and surfers realize that it's not going to stay the page that sited viewers were looking at. No, indeed! Neither sighted or blind are likely to stay with the first one! But for the blind it may take longer! He may have wished to follow a link to somewhere else! But frequently a blind user gets bogged down with technical details of the screen reader, interpreting text which was delivered to the ears or fingers in a jumbled manner, and meanwhile entirely lose interest in something upon which the sighted user would have followed up. Frequent technical errors with the software can indeed present a formidible challenge to patience!
If I am fortunate enough that the page reads out without further prompting from me, I sit listening either to the relevant paragraphs and links, or I tab through from link to link, hoping I might perhaps find something of use, depending on whether the Web designer thought to put an alt-tag or address as the link. I want to find something I might immediately determine looks interesting. I might tab through until I find a link I wish to notify someone of (or to pass along). The important point here is that this link must be clearly highlit, because if I use the Send option of the File menu the 'Send Link' or 'Copy this link location', is going to be grayed–out, i.e. inactive. The problem with this is that JAWS or Operating System is not intelligent enough to ask me what I am trying to do. This is a question in this case I could quickly answer: Send an identified link to a recipient. As a blind person, I cannot verify focus or have the luxury of copy and paste in the location field is not always available; neither does one always have active speech or avoid inactive. If as the blind user of screen reader software you need to read nearly every word on the page, it can take a lot more time, apparently, but a sighted person looking at the page for points of interest will jump intuitively to a word defining the subject in mind.
The counterpart of this for the blind user is a reformat of the page links into a column of links that descends alphabetically. One guesses what might be there and enters the first letter of what is desired. Then the focus jumps to a link in the list, and you must decide whether it is a link of interest. Say it's recipes. Then R will get it, but if you're after restaurants, not recipes, you type another R. If nothing comes up, you're out of luck. It is up to you to arrow up and down to find a link of interest, wondering whether any of the links in that list are relevant at all. If not, you close the dialog box in disgust. It becomes a hit or miss proposition. Shall we try entering a C for Cafes? B for bufets?
The blind user, then, needs more intuitive skill than the sighted user, because not everything logical or relational is presented to the ears in an orderly fashion.
To make tables easier to read every cell needs to be encoded with an identifier to indicate the row and column that it belongs to. Refer to http://www.santarosa.edu/access/checklist/example5.html#Description5 for examples of how to write the code.
Amy Cowen advises"3
"The WAI recommends that you use the ABBR attribute for <
User agents tend to read across the screen, so when you use a table design that puts columns of text on the screen newspaper-style; the sense of the content will be lost. If possible, provide an alternative location for a linear version of the content. (You can use CSS to position content visually and still have the source code be linear.) (end*3)
Filling Out Forms