This musician built an AI clone of her voice so anyone can sing as her
scientificamerican.com
March 3, 2026
4 min read
🔥🔥🔥🔥🔥
32/100
Summary
Holly Herndon has developed an AI voice clone that allows users to create music using her custom models. Her journey into machine learning began in 2015, evolving from initial "scratchy" outputs to sophisticated tools for musical expression.
Key Takeaways
- Holly Herndon developed an AI voice clone that allows users to perform using her voice in real time.
- Herndon emphasizes the importance of creating custom AI models and datasets to achieve unique sound outputs, contrasting with commercially available models that often produce average results.
- Herndon's work is described as "protocol art," where the creative process involves setting the rules and conditions for art generation rather than just the final media output.
- The AI models can generate music in the style of historical composers, allowing for collaborative performances that blend human interpretation with AI-generated outputs.
Community Sentiment
Positives
- Holly Herndon's proactive licensing of her voice model allows her to control access and potentially create new revenue streams, showcasing a smart adaptation to generative audio trends.
- By shifting the narrative from copyright infringement to creative collaboration, Herndon's approach could serve as a blueprint for other artists navigating voice cloning technology.
Concerns
- The rapid advancement of voice cloning technology raises concerns about artists' ability to control how their voices are used, highlighting gaps in current copyright law.
- Normalizing voice cloning as a participatory tool could change expectations around ownership and authenticity, creating tensions in the music industry.
