* Using Speech Dispatcher with Festival

It is recommended to run Speech Dispatcher with Festival, since Festival is the
speech synthesizer providing most functionality together with Speech
Dispatcher.  If you want to use Speech Dispatcher with Festival, please install
the speech-dispatcher-festival package and read its README.Debian.

* Using Speech Dispatcher with ALSA or OSS

When Speech Dispatcher is used with direct sound output to ALSA or OSS (on the
contrary to using a sound server, e.g. PulseAudio) and your audio device is not
set up to allow concurrent access, Speech Dispatcher may block all your audio
output.  For this reason the default Speech Dispatcher configuration comes with
PulseAudio as the only audio output.  If you don't want to use PulseAudio for
Speech Dispatcher audio output, change AudioOutputMethod option in your Speech
Dispatcher configuration (either /etc/speech-dispatcher/speechd.conf or
~/.speech-dispatcher/speechd.conf) accordingly and restart your Speech
Dispatcher processes.

* System wide startup

If you want to run Speech Dispatcher as a system wide process, set RUN option
in /etc/default/speech-dispatcher to `yes', and set the SPEECHD_SOCKET environment
variable to /var/run/speech-dispatcher/speechd.sock before starting clients.

* Log file permissions

Please note all Speech Dispatcher log files can reveal your user interactions
and texts you read through Speech Dispatcher, among others.  For this reason,
the /var/log/speech-dispatcher/ directory is world-unreadable by default.

 -- Milan Zamazal <pdm@debian.org>, Fri, 02 Jul 2010