OK, I did some debugging and found our a bit more information when trying to select voices for Cepstral using Speech Dispatcher. It seems that Speech Dispatcher is ignoring the keyword FEMAIL1 If I put Callie's name in the AddSpeech line in my cepstral-generic.conf line for FALE1, then Callie's voice comes up instead of David as originally configured. The synth command line shows this. The wierd thing though, if I use the spd-say command, it always respects the proper voice selection using the DEFAULTVOICE option from speechd.conf and the modules file. So I thought maybe there's a problem with speechd-up but I get the same problem when trying speechd-el for emacs so that would leave out speechd-up. So the fingers are now pointing at Speech Dispatcher. Are there any other ideas on what I could look for or is there a bug in the voice selection logic for Dispatcher? -- HolmesGrown Solutions The best solutions for the best price! http://holmesgrown.ld.net/
