I think William's suggestion of getting espeak to produce a temporary wave file and getting another program to play that might give better results on some systems. On my laptop the conf file I posted earlier gives me no problems on my laptop (except that it uses oss instead of alsa), but on my desktop that conf file seems to not allow silencing (and when using my Terratec DMX6fire card (ICE1712 driver), I also get very poor response) but by making the change to have another app to play the sound these problems disappear on my desktop. Also I guess it would be possible to get it to use alsa with that output system. My only concern with that is, with large amounts of text, you may get poor response as espeak will have to process all the text before the wave file is ready (I don't know if speech-dispatcher seperates large text into smaller chunks). also if you want to give espeak a different accent, you supposedly can, have a look at some of the docs about voices, dictionaries, etc, I think you are stuck with the voice itself though, as they haven't released the voice creation tools. There has been a little mention about it on the espeak forum. From Michael Whapples
