Mon Dec 05, 2005 4:43 am by Cold_Void
i wouldn't say it's a lot of hassle - i made a new Neuralnet voice in an afternoon with the XMLtoUTF&back converter + a speech synthesizer. but unless you like synthesized voices or you have a really nice suite for making them you'll be recording little scripted soundbytes for quite a while - the neuralnet has about 60 sounds it actually uses to give you an idea.
basically to do a GCS base,you have to record the base name in several voices to match,and add its $gcs name to the voices ini sections, then generate the _UTF_ hashcode/CRC dealie from the $gcs_base_name and prefix it with "0x",and finally import the sound files to each UTF node named with the UTF hex hashcode
Edited by - Cold_Void on 12/5/2005 4:49:23 AM