The patented 3Dmsg™ technology takes
advantage of a hybrid neural network and hidden markov modeling
speech technology to analyze spoken data and break it into components
that can be represented visually. These components can be thought
of as “visual phonemes” or “visemes” and represent different mouth/teeth/tongue
positions and movements. By stringing these visemes together an
avatar can be controlled or rendered to represent a person or creature
speaking.
3Dmsg™ has the most accurate lip-synchronization technology,
and the visemes can be created not only from live speech but from
input text or a combination of text and speech for the most accurate
representation.
|
