Re: Speech Dispatcher 0.4 released

[Bill Haneman <Bill Haneman Sun COM>]

Hi Hynek,

Thanks for the information on the new Speech Dispatcher release.

Regarding gnome-speech and Speech Dispatcher - thanks for working on the
Speech Dispatcher driver for gnome-speech.  One concern which we had
with earlier Speech Dispatcher versions was the use of GPL instead of
LGPL, which would prevent us from bundling this driver with gnome-speech
(and might limit the extent Speech Dispatcher use in GNOME).

Though drivers such as this, which use interfaces rather than dynamic
linking, are possibly in a sort of "gray area" with respect to GPL
linkage, if the interface used is only available in a GPL version it
raises concerns about compatibility with non-GPL code.  Our policy on
accessibility modules (and GNOME libraries in general) has been that
libraries and service APIs that are intended for use by multiple clients
and applications should use LGPL rather than GPL.  There are of course
exceptions, but gnome-speech certainly wants to retain LGPL
compatibility (note that if gnome-speech were GPL instead, it would in
most cases be impossible to write drivers for proprietary TTS engines -
at least as I understand the issues).

Has Speech Dispatcher 0.4 modified the license to LGPL?  If not, would
this be possible?

[more comments below]

> I'm currently working on writing a gnome-speech output driver for
> Speech Dispatcher. It already speaks (using the Speech Dispatcher
> supported synthesizers) on my machine and you might expect more news
> on it from me in short time.
> 
> We are also working on the index marking (or callbacks) support. We
> already have such support in Festival, although the corresponding
> interface in Speech Dispatcher is still missing (I hope it will
> be available in the next release). If I understand correctly, this
> might help gnome-speech with one of it's big problems.

You are quite right that support for callbacks and end-of-speech markers
in festival is something gnome-speech has needed for some time.  Thanks
very much for helping work on this.

I continue to have some questions about whether it makes sense for us to
have two separate speech service APIs which appear to have similar
goals.  What synthesis drivers are available for Speech Dispatcher at
this time?  (Currently we have six drivers currently available for
gnome-speech).

I look forward to your gnome-speech driver work, and to exploring
whether we might be able to more closely integrate our projects in the
future.

Best regards,

Bill