[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
A number of people have suggested assigning different festival voices to
specific nicks in IRC. I've contemplated the idea a bit, and basically
felt that it wasn't worth doing. Until now.
I've been digging through the MBROLA page
(http://tcts.fpms.ac.be/synthesis/mbrola.html), and found a program
(ftp://suburbia.net/pub/proff/zen-festival.el) that uses MBROLA to do
speech synthesis in the ZenIRC & ZenICB (Internet Citizen's Band -- sounds
interesting, eh ?). It's written in some form of lisp.
The significant difference between it & speech.irc+speechd is that it does
use multiple voices -- assigning them to new nicks at random. I like the
idea. Now I just have to figure out how, and at what levels I want to
I also want to be able to use voices from multiple different speech
synthesis packages simultaneously (rsynth/festival/mbrola).
Ohh... /ctcp SomeNick voice
-- query a person about which voice they wish to have represent them....
A couple of you (Kyle & ali) were talking about doing translation.
I figure it could be done by attacking bablefish.altavista.com, as you
were saying, but also as you were saying, I don't think it'd be real
useable, because of the amount of time it'd take to do the translation.
But the biggest problem I see is that of determining the source language,
so that you can decide what language to translate from. I don't think we
can. But you could set a language for a channel. So, if there's a
channel called #spanish, where everybody speaks spanish, you could set
everything from that channel to be translated from spanish to english (I
think babelfish can only translate to/from english).
I don't know if any of you have actually looked into MBROLA. It contains
about 12 free, multilingual speech synthesizers, of which Festival is one.
Gas powered skateboard that can do 30mph. I want one.
PGP fingerprint = 03 5B 9B A0 16 33 91 2F A5 77 BC EE 43 71 98 D4
email@example.com / http://www.op.net/~darxus
Far Beyond Reason