Is it possible to put a small hook in the unix code to just run /usr/bin/festival --pipe with the text exported to it? I see the unix code file is empty for this, and its a simple fix.
Festival is on just about every distribution at this point, perhaps just have it be a shell script you pipe into, this would allow users to change to whatever they want if its command line capable.
-- cut here --
echo $@ |festival --tts
-- end cut ---
And you will need to add this to the ~/.festivalrc file
-- cut here --
(Parameter.set 'Audio_Command "aplay -q -c 1 -t raw -f s16 -r $SR $FILE")
(Parameter.set 'Audio_Method 'Audio_Command)
-- end cut --
You can then send anything voice.sh <this is whatever>
and it will play back using Alsa driver.
The key is using a setable paramater to run the text, perhaps have to types
outputpiped|voice.exe and voice.exe ARGS.
Then most unix users can use what they would like at that point.
I'll open a pipe to a program, default to festival, and have it be an option in the .ini file if people want to change it.
Woops I'm a dumas
The festival --pipe I can do.. I see some prefer "flite" though. Hmm. And Speech Dispatcher would probably be the "most correct" choice, but it doesn't look like all distributions have that yet.
Looking at edgy eft, I see the festival package is rather mature and seems to work out of the box (and was already installed on my system), whereas speech-dispatcher seems to be a hack-in-progress :)
Could you do me a favor and check if either festival, flite or speech-dispatcher exists as vendor-supported packages for other distributions?