Oh. Sorry, I misinterpreted you.
That is a little more complex.
You need to record one wav file for each language ID and then assemble
them all using the voicefont tool provided on the VoiceBuilding wiki
page. It is quite a big job; you will need to record 402 wav files with
the current lang file. There are slightly more automated ways of doing
this, but they require coding which hasn't yet been done.

Does anyone know if there have been any license issues with the use of
Eloquence/Viavoice voices?

Jamie

On Tue, Apr 25, 2006 at 12:33:02PM +0200, Rainer Jost wrote:
> Hi Jamie,
> thanks for Your answer. I'm now thinking abot working on my own voice for that.
> Rainer

-- 
James Teh
Email: jamie@jantrid.net
WWW: http://www.jantrid.net/jamie/
MSN Messenger: jamie@jantrid.net
Jabber: jteh@jabber.org
Yahoo: jcs_teh