I completely disagree with the authors hypothesis that Kinetic Typography helps the reader, but damn that is an impressive application.
According to the video it automatically chunks Japanese text into words (which is nontrivial, since there are no spaces) and aligns both words and syllables with song.
Singing sounds different than normal speech and all voice recognition I have seen uses speech as data, so it is also non-trivial to align.
On top of that there's a (relatively) slick user interface.
According to the video it automatically chunks Japanese text into words (which is nontrivial, since there are no spaces) and aligns both words and syllables with song.
Singing sounds different than normal speech and all voice recognition I have seen uses speech as data, so it is also non-trivial to align.
On top of that there's a (relatively) slick user interface.
Bravo!