Voice Actor Yūki Kaji Counters Unauthorized AI Cover Songs by Offering Official Voice Synthesis Software

The rise in AI generated content has brought to light questions regarding the rights to celebrities’ images, likenesses, and voices and how we produce original works. While the debate is far from settled, the 2023 SAG-AFTRA strike helped bring the issue to the public. In Japan, the issue is looming in the background, but there’s a feeling no one really wants to address it. However, voice of Eren Jeager from Attack on Titan, Yūki Kaji, is trying to protect his voice with his new official voice synthesis software.

The X (formerly Twitter) accounts for Soyogi Soyogi and Yūki Kaji announced the project on September 3-4, 2023. The post by the Soyogi account noted that Kaji was embarking on a new project related to his voice to commemorate his 20th anniversary as a voice actor. In turn, Kaji posted how he is working on a voice synthesis project and would be updating regularly.

From there, information about the project was slowly revealed, which included the character design of Soyogi Soyogi and interviews regarding the project. On January 31, Soyogi Soyogi’s X account announced the opening of the official website.

Once again, information about the project was sporadic. Nonetheless, Soyogi’s X account posted new songs using the voice synthesis software semi-regularly. But on April 3, it was revealed a crowdfunding campaign would begin through Campfire to bring Kaji’s project to life.

The crowd funding campaign began on April 11 and ran until May 31 on Campfire. The initial goal was 10 million yen (about US$63,820), but the campaign raised over three times as much from 907 backers. The website notes the purpose of the campaign was to allow creators a method to use Kaji’s voice to create new and interesting stories and songs. Of course, implicit in the message is that he is providing legal avenues for fans to use his voice.

Prior to the crowdfunding campaign ending, Livedoor News reported on Kaji’s thoughts regarding generative AI, especially in the audio sphere, on May 28. Kaji states he’d been worried about copyright issues surrounding voices. Hence, he started the project so his voice wouldn’t be abused. However, Kaji went on to say, “Still, I believe we should coexist with AI, not be hostile to it. There is no good or bad in the technology known as AI. It all depends on the morality of the people who use it.” We can see he want to give people the tools to use his voice, but in a legal manner.

Kaji also posted on his X account upon completion of the crowdfunding campaign his gratitude to all those who backed it.

This ione sure fire way to help prevent the illicit use of Kaji’s voice. Hopefully, the Soyogi Soyogi program will become available to the public so Kaji’s fans can create all sorts of interesting music and other audio projects.

Update: Replaced trademark name with “voice synthesis.” Thanks, WeebRunningWild.

Sources: Soyogi Fractal’s website, Soyogi Soyogi’s X/Twitter account (link 2, link 3), Yūki Kaji‘s X/Twitter account (link 2), Soyogi Fractal’s YouTube channel, Soyogi Fractal’s Campfire crowdfunding campaign website, Livedoor News via Hachima Kikо̄