Hi Josh,
Thanks again for corresponding with me regarding this proposal. I actually
have not had a chance yet to get my hands dirty with espeakeditor, or any of
the voice-building components in the interface, for that. I do understand
that it is something that can be done with a standard PC running a *nix
OS, though the memory demands might be an issue. On the other hand, I would
hope that hardware limitations at OLPC only become less restrictive with
time. And of course, more considerable processing power might be had via
solutions like those offered by the OLPC School server project.
On a different note, your elaboration of the target end-user groups sums up
the topic very neatly! Also, I will be contacting the OLPC group in the
Solomon Islands for more on their TTS efforts.
Will you be doing any mentoring work during GSoC 2008?
Best,
Alex
On Sun, Apr 6, 2008 at 11:09 PM, Joshua Minor [EMAIL PROTECTED] wrote:
I haven't tried to make a new espeak voice either. It would be great if
there were more female and child voices in addition to the existing male
voices in espeak. I suggest you give it a try. If it takes a lot of ram or
cpu time or just a lot of steps for the user, that may limit the
possibilities - or just shape the focus of the project.
Your proposal for making voices easily probably appeals to at least three
groups. First to educators and developers trying to add new voices for a
particular language or country. The second group, is obviously kids who
might like to learn about voice synthesis or just have the thrill of
customizing the laptop to have a new voice. The final one is the disabled
community who would like to use the XO as a tool to help folks communicate
more easily.
I know that the Solomon Island OLPC deployment is interested in creating
voices for the local languages there. (
http://lists.laptop.org/pipermail/devel/2008-January/010412.html ) You
might want to contact them to see what sort of effort they have made and
what parts they found difficult.
I built Speak because there was no gui for espeak on the XO. It was a
pretty easy thing to make and seems to fill the gap nicely. The more people
put effort into the underlying synthesis engine and more ways to access it
(like the excellent speechd effort) the more powerful the system will
become. Adding more voices will be a great way to expand the appeal of the
system.
-josh
On Apr 5, 2008, at 8:15 AM, Alex Escalona wrote:
HI Josh,
Thank you for voicing your support! It's great to hear that there is
general interest out there for this type of activity.
I have to confess that I have not tried the existing process for adding a
new voice. However, I am aware of the efforts required to undergo such an
undertaking, and hope to make such endeavors more accessible for XO users
and their communities.
Can you share any experiences or knowledge that you might have on this
subject? I understand that you were involved in creating and maintaining the
Speak activity on the XO. As well, I have noted some interest in this
proposal on the devel list (e.g.,
http://lists.laptop.org/pipermail/devel/2008-February/011076.html). And of
course, I know that Hemant Goyal has done considerable work in forwarding
Speechd on the XO as a speech synthesis interface, as well as advancing
efforts in TTS in general.
Best,
Alex
On Fri, Apr 4, 2008 at 10:07 PM, Joshua Minor [EMAIL PROTECTED] wrote:
This is an awesome idea. A couple of people have contacted me to ask
how to add new voices to Speak. It would be great to make this process
easier.
Have you actually tried the existing process for adding a voice?
-josh
On Apr 4, 2008, at 5:58 PM, Alex Escalona wrote:
Hi Everyone,
I just created a page on the OLPC wiki detailing my activity proposal--Your
voice on XO http://wiki.laptop.org/go/Your_Voice_on_XO. I hope to
develop this activity via GSoC 2008. A brief abstract of my proposal
follows.
This is a proposal for the creation of a new activity for the XO that
would advance localization efforts in TTS development, as well as promote
the involvement of the local community overall. Your voice on XO would
consist of a long-term, community-based project to build and/or further
development of a synthetic voice for the language used locally (for more on
synthetic-voice building, see http://www.festvox.org/bsv/p710.html, and
http://espeak.sourceforge.net/add_language.html).
This activity would entail integrating the voice-building capabilities
of eSpeak http://espeak.sourceforge.net/, or perhaps
Festivalhttp://festvox.org/festival/,
into Sugar on the XO, as well as working to facilitate synthetic-voice
building in a classroom, or community setting (for an overall view of how
the voice building process might proceed, see
http://www.cstr.ed.ac.uk/emasters/summer_school_2005/tutorial3/tutorial.html
).
Your feedback and comments are much