[EMAIL PROTECTED] (Andreas J. Koenig) writes:
> I just registered for the module list the namespaces
>
> Festival::Client
> Festival::Client::Async
> Speech::Festival
> Speech::Festival::Synthesiser
>
> I'm not quite happy that we now have Festival in two different places
> and would welcome a great unification. Basically we can go either way.
> Speech is a good toplevel name but Festival is a famous library in its
> own right.
Yes, I was wondering about this myself when trying to decide what to
name my module. One thing to note is that all of the modules listed
do very similar things. I wrote Festival::Client::Async because the
other modules did not appear to be well-suited to my needs (meaning,
for the most part, non-blocking operation), and in this sense I'm to
blame for not consulting with the other module authors about adding
features to their existing modules.
Looking more closely at Speech::Festival, it seems that it would not
be too hard to merge Festival::Client::Async with it, as they have the
same goals, though the interfaces and internals are different (I
happen to think that mine is a better implementation of the protocol,
but that could just be Hubris)
I'm not sure to what degree Speech::Festival and Festival::Client are
maintained, perhaps they can answer here :)
> I'm not intending to push anything here, I just discovered that all of
> you were unregistered all the years and I'd like to apologize for that.
Thanks for getting the discussion started, it is definitely something
that needs to be sorted out.
There are a few namespaces in which Festival related modules can be
put. I do think the Festival:: namespace should not be used, as it is
not really very descriptive. On the other hand, Kevin has got the
FestVox:: namespace for things related to building voices for
Festival, so in that sense it perhaps makes sense.
I would prefer for the module name Speech::Festival to be used as a
namespace for interfaces to the *internals* of Festival, and have the
(new, unified, all-singing, all-dancing) module which talks to a
Festival server be called something else.
I also very much like the idea (which Richard has implemented) of
having a common Speech::Synthesizer interface for people who just want
to do text-to-speech and don't care about being able to evaluate
arbitrary Scheme expressions and fiddle with utterance structures.
--
David Huggins-Daines | [EMAIL PROTECTED]
Toolsmith | http://www.cepstral.com/
Cepstral LLC | We Build Voices
