1 MBROLA based American English female voice for the festival speech
4 This voice provides a American English female voice using the MBROLA
5 synthesis method. It uses a modified CMU lexicon for pronunciations.
6 Prosodic phrasing is provided by a statistically trained model using
7 part of speech and local distribution of breaks. Intonation is
8 provided by a CART tree predicting ToBI accents and an F0 contour
9 generated from a model trained from natural speech. The duration
10 model is also trained from data using a CART tree.
12 This voice can be activated via (voice_us1_mbrola)