OK I figured out that Mumble was perhaps overriding the volume settings. There's a slider in Mumble that lets me adjust the test-to-speech audio volume. So that is no longer an issue.
Still, I miss having the ability to type "spd-say" and having it work. It seems that a special speech-dispatcher process gets launched by my login user when Mumble is launched, and somehow that speech-dispatcher process is specific and only visible to Mumble. And this speech-dispatcher process doesn't die on its own. I used to have this in my .zshrc: SPEECHD_SOCKET=/var/run/speech-dispatcher/speechd.sock export SPEECHD_SOCKET This was with the Squeeze speech-dispatcher package, where the speech-dispatcher was run globally from /etc/init.d. Access to the speech-dispatcher socket was protected by the audio group I think, which limited access to /var/run/speech-dispatcher/speechd.sock to only those members in group audio. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
