[hts-users:00049] lip synchronization with cmu voices

["Nicholas Volk" <nvolk@xxxxxxxxxx>]

Hi,

I'm wondering how extract phoneme boundary times from Festival-based
CMU_*_HTS-voices for an animated cartoon character.
The segment boundaries calculated by Festival
are saved to the temporary lab file which is then processed by the HTS
engine. However, if I understood correctly the hts_engine contains
it's own duration model. I'm wondering if it is possible
to get the segment boundary info from it.
I reckon it's possible since the japanese HTS voice is apparently used in
the Galatea project.

regards,
  Nicholas Volk