Question about speech-dispatcher inclusion

[Michael Curran]

Hi all,

I have just started using Ubuntu as of a few days ago.

I have been using Debian and Gentoo (all with speakup) for many years, but 
this is my first time at being able to simply install a Linux OS all with a 
working and accessible GUI.

Congrats to the Ubuntu Accessibility team so far for making this a reality. 
And of course well done to the Orca team for making an ever improving screen 
reader for Gnome.

As the subject line says, I have a query about speech-dispatcher's inclusion 
in Ubuntu.

Is there a reason why speech-dispatcher can not be included on the Desktop 
live cd? I have tried both 6.10 and 7.04 and this seems to be the case.

I absolutly love ESpeak (for many reasons). But putting in the 7.04 live cd 
and having Orca come up talking with ESpeak the way it did made mee quite 
disappointed to see ESpeak used in this state.

The problem was the responsiveness, and the jumpyness.

As I understand, Orca was using gnome-speech, with a gnome-speech driver for 
ESpeak.

I am not blaiming the author of that driver, I'm sure there must be 
limitations to gnome-speech to have it act in this way.

But, as soon as I installed Ubuntu successfully, I quickly installed 
speech-dispatcher, and told Orca to use that instead.

Straight away, absolutely wonderful. Using Orca was pretty much as 
responsive as using NVDA, or Jaws or Window Eyes or Hal, on Windows.

Because I have only just joined this list, and only just started using 
Ubuntu, I'm sure I have missed why speech-dispatcher isn't included on the 
live cd. But, I really do think that we must push for it to be.

Or another option is to write an Orca synth driver for ESpeak directly. But, 
so far, my dealings with gnome-speech havn't been that good. Even with 
festival its still a little jumpy.

Its fine to have an accessible operating system, but I always feel the most 
important thing is responsiveness. Using Speakup, you press a key things 
just happen, using NVDA or Jaws etc, things just happen. But trying to use 
Orca with a slow synth driver, its seriously like walking through fog. You 
know where you are if you move carefully and slowly, but you can't just run 
like the wind.

I know that Orca is much better than this, but using it with the espeak 
gnome-speech driver, it wasn't that enjoyable.

Another surprising thing was that when I changed to speech-dispatcher, 
because espeak is forced by speech-dispatcher to use aplay, I was able to 
play sound files in Gnome quite nicely, and still have Orca talking.

Where as with gnome-speech, I think that wasn't working, at least on my 
system anyways.

I am thinking of spending some of my time, when I'm not developing NVDA that 
is, to perhaps work on Unbuntu Accessibility if I can. Perhaps one thing I 
could do is look at writing an espeak driver for Orca, or perhaps try and 
work out the issues that exist in gnome-speech.

I would really like to help out in anyway I can.

I am currently writing NonVisual Desktop Access (NVDA) 
[http://www.nvda-project.org/] which is a free and open source Screen Reader 
for Windows.
BTW: NVDA can use ESpeak via its sapi5 driver, and this is certainly my 
choice of synth, when I don't have access to Eloquence/Viavoice.

The main reasons I started developing NVDA was because of the price of 
commercial screen readers in Windows, and also the fact that a lot of blind 
users were not yet ready to move to Linux. Of course I'd love for all blind 
people to be able to use Linux full-time, but I'm sure we all realise that 
that is not yet the case.

I figured that if I could prove that a usable screen reader for Windows 
(that is free and open source) could be written, then perhaps more blind 
people would put their trust in this type of software, and perhaps give them 
ore confidence  to try out Linux, perhaps Ubuntu.

Anyway, Thanks so much for all the people who have done such fine work with 
Ubuntu, and Linux Accessibility in general.

Mick