Synthesis of speech and song: commonalities and differences, style versus individuality
Takeshi Saitou, Japan’s National Institute of Advanced Industrial Science and Technology (AIST) Hideki Kawahara, Wakayama University
This seminar will feature two distinguished guests from Japan. Dr. Takeshi Saitou of Japan’s National Institute of Advanced Industrial Science and Technology (AIST) will present his method for Morphing audio from speech to song, by which a text recorded by a speaker can be reprocessed into a song, preserving phonetic realism, with melody and other attributes of singing added.
To follow, Professor Hideki Kawahara of Wakayama University, inventor of the well-known STRAIGHT algorithm for voice synthesis, will present a project on speech analysis and modification method based on an elaborated VOCODER and a morphing procedure. He will then moderate an open discussion of possible ways of handling singing/speaking style and individuality separately.