Thoughts on projects and an rfc.

[Kirk Reiser]

Hi Folks:  It has been awhile since we had any discussion of various
speakup projects and their status.  I thought maybe I should share a
list of projects I/we are working on and maybe give a little status
report/comment for each one.  This idea was initiated by a
conversation I was having with Steve Dawes on the reflector last
night.  We were talking about the availability or lack there of for a
program to do caller id anunciation when the phone rings.  Sort of a
cool idea in my opinion for using voice capable modems to do caller
id.  I'll put it on the list below.

I have noticed over time that we are starting to get quite a group of
coders here on the list.  If any of you are interested in getting
involved with any of these projects it will certainly be appreciated.
Our team of active developers is quite small and so our projects don't
make a lot of headway very quickly.  Please remember you don't
necessarily need to be a good programmer or one at all to be involved
in these projects.  All projects have other rolls which are required
for the completion and availability of them, such as web page writing
and maintanance.  Cvs coordination and documentation are two other
examples of tasks which are important to any project.  Of course, no
project ever reaches the light of day without coders dedicated to
seeing the project through to fruition.

speakup speech console for Linux
This is probably our most complete and visible project.  Currently we are
working on a number of aspects.  Rewriting portions of the code so
that it will meet the linux kernel developers approval for inclusion in
the standard kernel tree.  I have been a bit lax on this because of
other responcibilities and projects.  Driver modulization is currently
under active development.  This will allow us to load and unload
synthesizer drivers at will so that we can finally include drivers for
devices which are either farily large or require resources unavailable
at the early stages of boot-up such as tuxtalk or a Dectalk-pc
driver.  Cut and paste between consoles is scheduled as the next task
after modularization, it will make it possible to mark sections of one
console to be copied over to another console.  The reasons for this
are things like copying a html URL or text from one system to
another.  After that a Dectalk PC driver so that we can provide linux
as a viable alternative for people which use that synth as their main
access to computers.  There are many other items on the todo list but
those are the main ones on the go currently.  More hands would of
course make these get implemented a lot faster.

awesome professional audio editing system for text based consoles
This one is sort of on hold currently.  Not because it isn't a
worthwhile project but because there are a number of other
programs/packages which have potential to provide the same basic
capability.  Frank and Jeromey have been working with Paul Davis on
Arder and I have been playing with the idea of a ncurses interface for
ecasound.  This is one of those projects that everyone uses an excuse
for why they are not quite willing to give up Windows because of a
lack of availability.  Kind of funny that there aren't very many
blinks actively working on it though.

EBTAFS port back to Linux Electronic Braille Translation and Formatting system.
This one is sad and very stupid.  Ebtafs is my braille translator
which I originally wrote back in the 80's with help from at least one
other active speakup user, for doing braille translation.  It is a
textbook style translator which provides automatic contents building,
volume splitting and hand inclusion of nemith code or foreign
languages.  Darcy ported it over to linux last fall but though a
stupid oversight on my part we lost it all when the automated
accounting system on speech removed his account before I realized what
was going on.  Sorry Darcy, this bothers me a lot because of all the
hard work you put in.  Needless to say, it all needs doing again.

OCR system for Linux
This one has also gone through a number of revisions.  It is the other
reason people always give for why they are not quite willing to give
up on Microsoft.  We originally found a doctoral project which we did
a large chunk of work on to bend it to our ways.  That work is still
available in cvs under the name socrates.  Meanwhile back at the rest
of the world project, there is another group of people actively
developing a system known as gocr/jocr.  Their work is comparable to
socrates and in some case moving forward faster than we can move
socrates forward.  We are now working on a ncurses interface to
libgocr and will probably give socrates a lot more flexibility and
power in the long haul.  It also has the considerable advantage of
having a number of people working on it around the world.

html preprocessor for ebtafs
This one isn't even started because of the lack of availability of
ebtafs.  It is basically a program to convert html tags into ebtafs tags.

xml browser and player for new digital talking book standard
The national libraries services of the government and the daisy
consortium have been developing standards for a new digital talking
book format.  A number of organizations around the world are coming up
with commercial devices and software packages for windows.  This
project is to develop a gnu project for linux.  I have one person
currently actively working on it.  He is still a student and not blind
so he is strugling with the concepts of blindness issues and reading
other peoples code.  He is making progress though and hopefully will
get things pieced together with code at some point in the future.  The
standards are not exactly trivial to get your head wrapped around
technically anyway.

command line real audio player without X
This project was taken on by Matt Campbell and he's done a wonderful
job of it.  It is called trplayer and many of you are familiar with it
already.  There is a mailing list and a web page and Matt's the man to
ask about any details or ideas.

projects web page with subpages for each project.
Well, this is probably selfexplanitory and doesn't exist yet.  We
would like to have a page for each project with the projects current
status, who's working on it, how to get involved and a number of
issues.  This is a good example where programming expertise isn't
necessary, but desire to help and a good work ethic are necessary.

software synth good quality based on rsynth
It isn't currently what I'd call good quality, but it is working and
called tuxtalk.  It is in cvs for anyone wanting to play with it.  It
still needs a lot of work and is not a trivial project.  Speech
synthesis is state of the art so the field is moving forward faster
than the packages can keep up.  This one has an emphasis on smallness
because it is scheduled to be the first software synth for speakup so
it needs to be loadable as a module.  We have already modified it to
accept in stream commands based on the DoubleTalk.  It currently
supports rate and pitch change, punctuation levels and most important
instant shutting up.  We need someone to take on contacting all of the
original authors of rsynths pieces of code so that it can become a
bonifide gnu project.  Of course once again we need coders interested
in speech synthesis.

Speak Freely modifications and enhancements
Our favourite internet audio package is speak freely.  It allows
people to talk and type/chat over the internet in realtime.  It has
fairly good audio quality and is available for linux and windows.  We
use it as our speakup reflector where people get together to help each
other, ask questions, do project design, general bullshit and gossip
about people not on the reflector! 'grin'  If you are interested in
getting involved in a speakup project it probably would behoove you to
set it up and get on the reflector.  We have made a number of
significant advances in it's feature set and are continuely working on
it.

Caller id daemon
I thought that Steve's idea was cool, so I am making a project for it
at least in name.  This would use a voice capable modem to monitor the
phone line and announce the number and maybe other information when
the phone rings.  It would then blast that information either to the
console or as synthetic audio to advise the user of who's calling.

There is one other project which is under active development that I
cannot announce openly here.  Many of you are already familiar with it
through your active participation.  If you are just beside yourself
with curiosity, you can either get on the reflector and ask or find
someone else involved and ask them.  It's worth it.


If you have other projects you would ike to work on and would like
them part of the speakup project, post them and let us have a look.
If you would like to get involved with any of these projects greate!
We have way more projects then we have people working on.

If you are just interested in being an armchair designer, you might as
well piss-off.  There are very many exciting things to work on and
getting should'd to death just generates frustration.  Get involved
and help develop packages for the blind community or the linux
community in general.  It will make yu feel good about your part in
the overall machine and will make others appreciate you as a
contributor.

Once again, I wish to thank all of the folks that have in the past and
those still working on the various projects.  I couldn't do it without
you and would certainly not be as happy not getting to know you.  We
make significant advances as a team and it's always fun when it works.

  Kirk