getting started wit h speech-dispatcher

[DON . RAIKES]

Hi all,

I just installed speech-dispatcher 0.6.6 on my enterprise linux 5.0 system.

I started the server, and can run spd-say and it talks just fine.
I also have ibmtts installed.

The particular kernel I am using has speakup built into it, so I am wondering 
how I get speakup to work with speech-dispatcher.
I did compile and install speechd-up as well, but haven't been able to figure 
out from the docs how to make all the pieces work together.

Any tips would be appreciated.

-----Original Message-----
From: Halim Sahin [mailto:address@hidden
To: address@hidden
Sent: Monday, April 14, 2008 10:44 AM
To: speechd at lists.freebsoft.org
Subject: Re: Alsa/Pulseaudio conflicts


Hi,
On Mo, Apr 14, 2008 at 05:48:10 +0200, Hynek Hanke wrote:
> 
> Hello Gilles and others,
> 
> > After the Ubuntu speech dispatcher package has been installed (currently 
> > system-wide), there is no sound at next login: sd uses alsa which 
> > prevents a correct start of the user's pulseaudio session (pa 0.9.10).
> > I guess that distros could provide a new sd package for avoiding this 
> > conflict, e.g. :
> >   
> In principle, this is only a problem if this system-wide Speech
> Dispatcher uses ALSA, not PulseAudio directly. It should
> be possible to set it up to use PulseAudio in such a way that it
> doesn't block any other PulseAudio applications nor Speech Dispatchers
> running inside user sessions. There is also no problem if the
> user doesn't run PulseAudio at all. I agree that distributions must
> make some decision about it and should provide a setup
> that works, because mixing these two things brings troubles,
> as you described.
 
Ack mixing these apps isn't a good idea.


> PulseAudio is a necessary and good way how to finally
> solve the problems we are constantly facing with audio
> output. 

NACK.
Please read my latest mail!!!!
It is an additional layer between alsa and speech-dispatcher.
It will never be as fast as the current speechd alsa implementation.
The reopening problem is a good startpoint but I couldn't find another
app which works without a delay.

> One such example is the independent controll
> over the volume level of different clients -- in other words,
> raising the volume of Speech Dispatcher voice output,
> a repeated requirement currently impossible to accomplish

Ok that might be useful.

> with the architecture of just Alsa. But of course this is only
> one of the numerous technical advantages of having an audio
> system.
 
This is def. the only thing (volumecontrol) what makes sense.
In my opinion we should use alsa's softmixing instead of adding another
complex layer!!!!


>  From the standpoint of a developer of Speech Dispatcher who
> has some non-trivial knowledge about the various sound susbsystems,
> I can tell that the current direct ALSA sound output in Dispatcher is
> very limited, has problems that are non-trivial to fix and we don't
> like it. PulseAudio could be a huge improvement.
> 
Sorry Have you even tried to work with a screenreader using speechd and
any form of soundserver which works fast as expected?



> The current implementation of Pulse Audio in Speech Dispatcher
> is slow because the implementation of it contains a problem.
> For each sound output, it opens and closes the whole connection,
> so there is a big overhead which is completely unnecessary.
> We believe PulseAudio will be much faster once this is solved.

???????????
BR.
Halim

> 
> _______________________________________________
> Speechd mailing list
> Speechd at lists.freebsoft.org
> http://lists.freebsoft.org/mailman/listinfo/speechd

-- 
Halim Sahin
E-Mail:                         
halim.sahin (at) t-online.de

_______________________________________________
Speechd mailing list
Speechd at lists.freebsoft.org
http://lists.freebsoft.org/mailman/listinfo/speechd