Actually, I don't think you're doing anything wrong at all. I think the problem is in the systemd service file, and it's all my fault. I see I'm not even taking arguments from /etc/conf.d/espeakup as I should be doing. I am a bit unsure how to take arguments from this file when running from a systemd unit, so any patches to /usr/lib/systemd/system/espeakup.service are definitely welcome. As this file is Arch specific for now, patches against this file can probably be submitted via the bug tracker. Otherwise, as I sent the original service file to this list, I don't think there will be any problem sending patches to it the same way. Since Tom Gundersen added my file to the package, he is probably the best person to discuss how patches or rewrites should be submitted. Meanwhile, I'll research how to pull command line arguments from the config file and incorporate them into the service, and if I can do this, I'll submit the line to add or a patch here in this thread. ~Kyle http://kyle.tk