Annotation of pkgsrc/audio/festvox-en1/DESCR, Revision 1.1
1.1 ! zuntum 1: MBROLA based British English male voice for the festival speech
! 2: synthesis system.
! 3:
! 4: This voice provides a British RP English male voice using the MBROLA
! 5: wave synthesis method (via the MBROLA package). It uses a modified
! 6: Oxford Advanced Learners' Dictionary for pronunciations. Prosodic
! 7: phrasing is provided by a statistically trained model using part
! 8: of speech and local distribution of breaks. Intonation is provided
! 9: by a CART tree predicting ToBI accents and an F0 contour generated
! 10: from a model trained from natural speech. The duration model is
! 11: also trained from data using a CART tree.
! 12:
! 13: This voice can be activated via (voice_rab_mbrola)
CVSweb <webmaster@jp.NetBSD.org>