Re: [orca-list] espeak-extra-voices

[Kyle <kyle4jesus gmail com>]

According to Jonathan Duddington:
# The voice files are scripts which are interpreted by eSpeak at runtime.
#
# If you make a new voice file, its licence (if any) is not determined by
# the licence of the eSpeak program.
#
# If your new voice is derived from voice files in the eSpeak
# distribution, then in theory it should respect the licence of the
# original voice files.  But voice files are trivial.  The contents of
# the language voice files are determined by their function and are
# therefore not copyrightable.  Even the choices of pitch and formant
# values in the voice variant files may not be, although I'm not sure
# about the legal position on that.

So then if compiling a bunch of these voice files into a single archive
and/or repository, GPLv3 should be safe enough, since it is the license
already used by eSpeak. Personally, I would be in favor of waiving
copyrights on voice files, assuming that is even necessary, but it's
more logical in this case to cover the compilation of voice variants
under the same license as the speech synthesizer that uses them.

# A better solution would be for Orca to allow the user to specify both a
# language voice and a voice variant, which is what the NVDA
# screen-reader does

I fully agree with this. I'm just attempting to solve a long-standing
problem by means of an easily installable work-around until such time as
Speech-dispatcher and Orca recognize language files along with a
separate list of available variants in the same way that NVDA does.
NVDA's method of choosing eSpeak voice variants is certainly more
logical than what we currently have, but anything is better than nothing
untill speech-dispatcher and Orca do work as expected.
~Kyle
http://kyle.tk/
-- 
"Kyle? ... She calls her cake, Kyle?"
Out of This World, season 2 episode 21 - "The Amazing Evie"