Hi all,
I was wondering whether it is possible to manipulate the parameters of a voice trained with HTS, and if so, how to go about doing this. I know that it is possible to manipulate the pitch contour or the duration of an utterance using Praat, which resynthesizes the utterance, but for doing something very simple and global, such as assigning a completely flat f0 contour to all output utterances, is there a way to do this by manipulating the trained models, or perhaps at the synthesis stage using hts_engine?
Thanks,
Erica