Let the ‘Perfect Voice’ Always Be There, Positive Affirmation and Negative Anxiety from Online Commentary

The application of artificial intelligence (AI) technology in the field of music industry is a growing phenomenon worldwide. The recent emergence of Al singer trained by users on video platforms and social media has become an important topic of public discussion in China. Based on interviews with users who train AI singers, this paper discusses two acoustic models Sovits 4.0 and DiffSinger for training AI singers. Additionally, to make AI technology adapt to human well-being, we extensively explores the positive affirmation and negative anxiety from online commentary about the ethics involved in AI singers, which might genuinely enhance critical understanding of AI musical ethics. By analyzing representative cases and collecting users’ comments about AI singers, human singers’ rights of personality and the instrumentality of their voice have been discussed.