An Update about Speech Synthesis for Sugar

[Hemant Goyal]

Hi,


> * How many languages does speech-dispatcher support?
>
> http://www.freebsoft.org/doc/speechd/speech-dispatcher_5.html#SEC8
>
> SD works with Festival http://www.cstr.ed.ac.uk/projects/festival/
> English, Czech, Italian, Spanish, Russian, Polish...
>
> What is the mechanism for adding additional languages? Phoneset
> recording, dictionary, and what?


As of now we are planning to use the eSpeak  module with Speech Dispatcher.
It is already available on the XO, is lightweight and has support for many
more languages as compared to Festival, and I quote from the eSpeak website
:

eSpeak does text to speech synthesis for the following languages, some
> better than others. Afrikaans, Cantonese, Croatian, Czech, Dutch, English,
> Esperanto, Finnish, French, German, Greek, Hindi, Hungarian, Icelandic,
> Italian, Lojban, Macedonian, Mandarin, Norwegian, Polish, Portuguese,
> Romanian, Russian, Serbian, Slovak, Spanish, Swahili, Swedish, Vietnamese,
> Welsh. See Languages <http://espeak.sourceforge.net/languages.html>.


I did some preliminary research regarding the language support that was
needed wrt locations where the OLPC Pilot Projects will be launched or are
running.

Austria - German
India - Hindi
Spain - Spanish
and so on..

For a full list of languages and how Languages can be added to eSpeak please
check http://espeak.sourceforge.net/languages.html. Jon SD the developer of
eSpeak has written a few tutorials about how one can go about adding or
improving the existing phoneme data sets. In fact Assim had documented and
played around with voice characteristics of eSpeak to improve the voice
output. You can check the results on the wiki page i had mentioned
previously [http://wiki.laptop.org/go/Screen_Reader#Voice_Files]

I have extensively used eSpeak through the course of this project, and will
certainly explore other supported TTS engines, however I am taking one step
at a time.

Now for organization.
>
> Where should we put TTS projects for language support? Can we create
> http://dev.laptop.org/tts? Who should be in charge? What sort of
> process should we have for creating projects? Should we just
> automatically create a TTS project for every translate project?


Indeed this sounds like a great idea, and I was planning to involve more
volunteers, once we have a basic speech synthesis frameworkpresent on the
XO. It will be easier to motivate individuals from different countries to
add translations/ improve phoneme data once there is "something" for them to
use. Again, I am a student and devote whatever time i get after my school
hours to this project, and hence the pace has been slow.

> speech-dispatcher is not packaged as an RPM for Fedora,
>
> I see Debian packages. Is there a converter?


It is possible to convert .deb packages to .rpm using "Alien". I am afraid
that wont be possible, as Fedora has strict packaging guidelines that we
must adhere to. I am writing the packaging rules from scratch. So it must be
approved by the Fedora community before it is accepted in the OLPC builds.
The reason for the strict guidelines is mostly Quality Assurance and proper
integration of the package with Fedora.

Have you looked at Oralux, the Linux distro for the blind and
> visually-impaired?
> http://oralux.net/
> We should invite them to join our efforts.


I did not know about this, and will surely go through there website. However
this is an extremely high-level decision that many developers from OLPC will
need to take, as I am sure there will be issues of integration of Oralux
technology with Sugar etc.

 So again I'll say I am moving one step at a time. Thanks for the points
that you put forward, they will help us to better structure our own ideas
and vision regarding tts on the XO.

Warm regards,
Hemant
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.laptop.org/pipermail/devel/attachments/20080219/fd1670ad/attachment.html>