Soyogi Fractal is a voice AI project built from Yuki Kaji's voice, with a live AITuber stream set for November 21, 2026 and a five-song mini album, Butterfly Effect, due the same day.
Soyogi Fractal is the name Yuki Kaji's voice AI project carries into two public-facing formats. The first is a contest: nine Yoyogi Animation Academy campuses will produce original CM narration by combining the Bon Soyogi voice synthesis software, built from Kaji's voice, with students' own voices. Kaji heads the judging panel, and the project's stated position is that using AI is not the goal, and the question is how to use it as an extension of expression.
The second format is a broadcast. Bonten Soyogi, Kaji's AI character, will appear as an AITuber in a live stream on November 21, 2026, starting at 16:00 on the project's official YouTube channel. The birthday program pairs the character with Kaji in real-time conversation, with game commentary planned and the game title withheld until the day. A second half puts Soyogi on stream alone.
Both stories put the voice AI in front of an audience without a fixed script. The earlier outings were staged, including a 3D live at Tokyo Garden Theater in March 2026 and a conversational app. The November stream and the same-day mini album, Butterfly Effect, are the next scheduled points for the project.
Written with AI from the cited Yomimono stories below, each itself sourced.
See our editorial policy. Every fact links
the story it came from.
Oct 27
Voice actor Yuki Kaji's AI character Bonten Soyogi will appear as an AITuber in a live stream on November 21, 2026, starting at 16:00 on the project's official YouTube channel. The birthday program pairs the character with Kaji in real-time conversation, with game commentary planned and the game title withheld until the day. The second half puts Soyogi on stream alone, and a five-song mini album, Butterfly Effect, arrives the same day.
Sep 11
Voice actor Yuki Kaji's voice AI project Soyogi Fractal and Yoyogi Animation Academy will hold a joint CM narration contest. Students at the academy's nine campuses will make original narration by combining the voice synthesis software Bon Soyogi, built from Kaji's voice, with their own voices. Kaji heads the judging panel. His quote: using AI is not the goal, the question is how to use it as an extension of expression.