Re: [orca-list] Proposal: Remove the emacspeak suppport

I agree here.  If you want a hardware synth, you can probably come up
with something comparable and much more cheaply by using a dedicated
low-end sound card for your speech needs and a nicer one for
multimedia.  Then, find some way to have pulse audio (since that's the
wave of the future) only aware of the multimedia card while the
cheaper one can only be accessed by Alsa propper so that you can get
speech without interruptions.

Another option would be for someone to write some sort of emulator
program that would make Linux think the sound card is a hardware
synth.  Then, something like espeak or flite could be embedded into it
somehow and, bingo, you have a hardware synth.  I can see where one
would be handy for things like musical programs that need to take over
your sound card with Jackd or with Speakup in console mode so that you
don't have to rely on something like Pulse Audio for the driving of
your speech device.

A third solution might be for someone to figure out a way for a cell
phone device running something like Android to serve as an external
speech synth.  That way, no matter where you go, you have your speech
solution with you.  A way to embed a portable Orca into that device
which turns on when it's connected to a given machine would be nice
too.  Might even make it a commercial venture since it's a
nonessential enhancement that would serve a niche group.  I'd sure pay
for it, for one.  Anyone working as a sys admin in a Linux environment
might too.  Especially now that accessibility for several desktops is
becoming more and more viable.  I'm reminded of how popular portable
NVDA is among people.

Just tossing out ideas.

Alex M
On 1/4/12, Thomas Ward <thomasward1978 gmail com> wrote:

I for one feel that the Emacspeak driver  support should be removed.
My primary reason for saying this is that very few computer users
actually own and use a hardware speech synth like a Dectalk,
Doubeltalk, Speakout, etc these days. A software synth solution like
Espeak, Cepstral, or Voxin/Eloquence  is far more portable and cheaper
to own and use. Plus on top of that most older hardware synths are not
even compatible with a modern PC anyway.

For example, I owned three different hardware synths and can't use any
of them on any of my computers. I have a Dectalk PC, but because it
uses an ISA bus it won't work on a modern desktop board that have PCI
bus slots. I have a Dectalk Express but being a serial device it won't
work on either of my laptops because they only have USB 2.0 ports. I
tried a couple of different USB to serial adapters and neither worked.
Finally, I still have my Braille N' Speak 640K, but again same problem
as the Dectalk Express. No serial ports and a USB to serial adapter
doesn't appear to work with speechbox mode.

In conclusion hardware support for older synths is getting less and
less viable, and the numbr of users still using them is rapidly
dropping off in favor of ESpeak, Eloquence, or something else. So
continuing support for the Emacspeak synth drivers really is
unnecessary at this point. I feel we could clean it out and use
speech-dispatcher for all our synth needs.

After all, if we want hardware synth support then we would be better
served by writing all new drivers for the current hardware devices. I
know there is a newer USB Dectalk Express around which should replace
the old serial model. The Trippletalk PCI and USB models are also good
candidates for new drivers.I'd much rather see effort put towards
supporting USB and PCI based hardware through speech-dispatcher rather
than hanging onto old ISA and serial hardware synth drivers.

On 1/4/12, Joanmarie Diggs <jdiggs igalia com> wrote:
Hey guys.

Back in June [1] I asked you all if anyone was using Orca with
Emacspeak, and as an aside suggested that it would be great if we could
have support for hardware synthesizers in speechdispatcher.

What resulted from my question was more of a discussion than a concrete
consensus. But the feeling I got from that discussion was that we have
people using Orca and people using Emacspeak, but we do not have people
using hardware synths with Orca via Emacspeak.

If these is indeed the case, and given the plans to make presentation
modalities "pluginable," having old cruft around which would have to be
converted is more trouble than it's worth *if* it's honestly not
something anyone needs.

So... Discussion is, as always, encouraged. But at this point it is a
proposal on the table for which I'm requesting a vote.

Thanks as always, guys!


orca-list mailing list
orca-list gnome org
Visit for more information on Orca.
The manual is at
The FAQ is at
Log bugs and feature requests at
Find out how to help at

orca-list mailing list
orca-list gnome org
Visit for more information on Orca.
The manual is at
The FAQ is at
Log bugs and feature requests at
Find out how to help at

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]