Anyone still interested?

[thomaslloyd]

I have been reasonable quiet on here for a long time now.

However work still continues, the latest a greatest developments
include:

* The vista TTS voice tested and working
* A new threaded speech system improved res[responsiveness
* Direct pulseaudio access increases performance over espeak
* Wav File generation
* Complete code rework
* Code efficiency testing to improve performance

I am not calling for testers as I did not want to waste peoples time
until at least an alpha release candidate is ready. I am posting as
things get imple,ented and a, happy to announce that the latest
performance result put open-sapi up there with the current free TTS
engine.

Would be good to hear from anyone still intrested as I am not sure
what you guys want / need a the moment. I know the release of Karmic
has been a bit of a disaster for speech users but if is it is any
consolation then open-sapi performace very well under karmic so watch
this space. Any feedback will help to spur me on as I feel as if am
developing this for myself at the moment.

Tom

[A]

Hallo,

I'm still very annoyed that my wife needs to use windows so still interested :)

Although time seems to become lesser and lesser over time I will be
happy to try open-sapi once it is thought to be stable. Or at least if
it is easier to find out what crashed it so I can report for you to
fix.

Cheers

[Tony Sales]

Hi Tom I think people are still interested in using SAPI5 voices as
the quality is much better then most open source voices - however I
think people want a working easy to install system and when it gets to
that stage I think interest will increase dramatically as people will
want to use voices they have legally purchased on Linux systems. The
best way to demo this would be for us to make a custom version of
vinux to demo it and although we can't distribute a configured version
we could post this on youtube etc which might increase interest.
Ultimately it might be possible to supply the program with a script to
download the sapi system and voices etc, or it might be possible to
set up a commercial distro which paid a percentage of each sale to the
licence holder like voxin do with eloquence - they only charge $5 and
offering a ready built system might give us an advantage over voxin.
On the other hand if you don't think there is enough interest then you
could start working on improving an open-source system or voice etc,
and if you ever want to get more involved with the vinux project you
would be very welcome as you have a much deeper understanding than I
do of the in depth software and hardware issues - I am just an
enthusiastic bubbling amateur after all! But don't tell anyone - they
seem to be under the impression that I know what I am doing!

drbongo

--

Dig that crazy beat on the drums: <http://vinux.org.uk> The best is
getting better!

[clayjar]

Tom,

Thanks for all your work.

I've been manually adding speech files (Japanese) on nmkendokai dot
com (Click on Kendo), but I would be highly interested in
incorporating OpenSAPI for dynamic TTS on the fly. I do a lot of
localization, and enabling speech for websites would also be a great
way to serve the public.

So, your work would be much appreciated. I have yet to find time to
pour over your work so far, but when I do find time in near future, I
will try this out. And I should point out that TTS is one of the
hottest tools for people involved in international or localization
work/education. I know few international dictionary sites that use
TTS in somewhat of a static way, and I think incorporating OpenSAPI
for a dynamic TTS on a dictionary site would be great.

Just looking at the list of features gets my heart pumping already.
X-) I'm eager to pour over the code and the documentation soon.
Thanks again!

Best regards,
M.H.

[thomaslloyd]

Well this is good news, any adoption of the project is always good and
all it needs is people to think how it can be implemented. As it
stands it is far from ready for use in every day life but it is coming
slowly with time.

One of the ideas was to create a very high level interface to get text
to speech working so that you do not have to be a programmer to get
text to speech working all you need to be is able to use to the
command line.

The integration into the web at the moment is a concept, by that i
mean fast on the fly text to speech generation by multiple clients, I
believe it is possible to connect the open sapi server directly to web
clients. As yet i have not managed to test this theory. At the moment
you can generate speech per request and serve it to the client quite
quickly but not directly you have to use files. I have my doubts over
how many simultaneous users open-sapi can handle as at the moment it
deals with one at a time. I have not spent time to make it a true muti-
user server but thats on the agenda next year.

It also is unknown how well it will support other languages and it
would be good to work on with people from other countries. I know
there are potential issues with how it is implemented at the moment
but i know it will work given the time and effort.

Tom

[Mitchellt]

Just wanted to offer up my encouragment. My wife is legally blind and
most accessibility options available today are both lame and
expensive. We use Ubuntu because it comes with a free screen
magnifier that is better than most expensive windows versions.

keep up the good work!

[thomaslloyd]

Thanks fr the support all, things are very busy in my life right now
and there is not another developer at the moment to take some of the
work on so it has halted for the past two months. January should see a
new injection of enthusiasm.

Hope you all have a nice Xmas and new years and hopefully we should
see a super dupa release in Jan sometime.

Tom