From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (qmail 3197 invoked from network); 10 Dec 1996 03:52:50 -0000 Received: from netcom21.netcom.com (jrebman@192.100.81.135) by mail2.redhat.com with SMTP; 10 Dec 1996 03:52:50 -0000 Received: (from jrebman@localhost) by netcom21.netcom.com (8.6.13/Netcom) id TAA16860; Mon, 9 Dec 1996 19:51:47 -0800 From: jrebman@netcom.com (Jim Rebman) Message-Id: <199612100351.TAA16860@netcom21.netcom.com> Subject: Re: another idea for speech To: blinux-list@redhat.com Date: Mon, 9 Dec 1996 19:51:47 -0800 (PST) In-Reply-To: <199612072244.AA07248@world.std.com> from "Lar Kaufman" at Dec 7, 96 05:44:29 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit List-Id: I will respond primarily to the comments made by Lar, but I will also answer a couple of others that have come up since his original post, and then, I hope this thread will end. Lar writes: > I can't agree with the idea that Synthavoice should be working on an > NT screen reader. IBM has already shown in OS/2 Warp 4 that it is > quite feasible and apropos for voice support to be an OS-supplied > interface, and I'd rather see pressure for Microsoft to adopt IBM's > solution to have standard hooks. If, when you talk about an OS-resident voice capability, you are talking about the voice I/O capabilities that are built into Merlin, then I have to assume that you have never tried to use this kind of speech to accomplish any kind of useful or productive work -- this kind of highly inflected, natural-sounding speech is totally inappropriate for a blind user. Because of its highly-inflected nature it is nearly impossible to understand at rates that are required for productive work; also, the amount of listening fatigue that results from prolonged exposure to this kind of speech would drive a normal person mad in short order. Microsoft and Sun, to name but two, are also developing these so-called speech APIs. Just fine for the bored executive who wants his e-mail read to him, or for the computer hobbyist who likes "neat, cool, and shiny" things. Of course, the kind of controls that one assumes to be in a normal screen reader are nowhere to be found in these speech APIs, either -- none but the simplest and most rudimentary. If, on the other hand, you are referring to the kind of hooks that are built into OS/2 that make it possible for Screen Reader/II to function, then you have missed the obvious point that Screen Reader/II *IS* a screen reader, and blind people who want to use computers need to use screen reader software to realize the maximum flexibility, control, and speed that it can offer over a simple, so-called "speech API". Just for your information, Microsoft has developed a standard set of "hooks" (actually, it is a much more powerful accessibility API), unfortunately, it was not mature enough to include in NT 4.0, however, it will be in the next release. You can check out the API for yourself, and even download the SDK -- go to www.microsoft.com/windows/enable/ There ye shall find what you seek. The bottom line is that whatever platform you decide to use, wether it be OS/2, NT, or X386, the only effective way for a blind person to work with it will be through the use of a screen reader, at least at this point in time. Then you continued: > (Let Microsoft support X? Why not?) Sure! Why not! Why don't we just hand over Linux, VMS, OS/2, and all the screen reader and other accessibility aid developers to them as well... You must be joking. I really can't see giving Microsoft any more influence then they already have (which is too much to begin with). No, No, No, a thousand (and twenty-four) times NO! Now, on to the other folks: To Mike Malver: Emacspeak will do what you want -- it won't work very well with ELM as it is "right-out-of-the-box", but with some tweaking it should do just fine. Switching between virtual terminals, each one running whichever shell you want, is a simple matter of hitting the alt-f1 through alt-f6 keys -- in the standard configuration of the version of linux that I am running, the default is for 6 virtual terminals f1 - f6. You can change this up to 250+ virtual terminals if you wish, but changing from one to the other is very simple. And finally to Kenneth... *WE* know what we are talking about -- SynthaVoice makes screen reader software, as well as screen enlargement software. They were the first company to develop a screen reader for Microsoft Windows (Slimware Window Bridge), and their current version is one of the finest, most stable, and powerful Windows/Windows '95 screen readers. They do not make speech synthesizers, and yes, I do know the difference between a screen reader and a speech synthesizer -- I've been doing this for a while. ------ SynthaVoice *should* concentrate on an NT screen reader -- it is something that is desperately needed right now. They don't need to spend any time, money, or effort on a screen reader for X, and next time I talk to Laslowe, I'm fairly certain I can convince him of this. Why should he make a huge investment in time and money, and compromise his already outstanding products, when the group I am affiliated with is going to deliver an X screen reader and give it away for free? It is a business decision, and a reasonably easy one, at that. -- JR