[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
add speaker personality information
The conventional way to implement voice
conversion is to find the mapping rules between the source and target speaker,
linearly or non-linearly.
However, I always thought that the mapping
rules could not precisely display the target speaker's characteristics.
There must be some crucial factors that help determine the perceptual feeling of
the speaker's personality. So, I wonder if there is a way to add the
speaker's personality information to a text-to-speech voice and how to
Would someone give some suggestions?
College of Signal Processing &
Nanjing University of Posts & Telecommunications
Jiangsu, People's Republic of China, 210003