SingVisio: Visual Analytics of Diffusion Model for Singing Voice Conversion
Liumeng Xue, Chaoren Wang, Mingxuan Wang, Xueyao Zhang, Jun Han,, Zhizheng Wu

TL;DR
SingVisio is an interactive visual analysis system that explains and visualizes the diffusion process in singing voice conversion, aiding understanding and comparison of different conversion conditions.
Contribution
The paper introduces SingVisio, a novel visual analytics tool that enhances interpretability and comparison of diffusion models in singing voice conversion.
Findings
Effective visualization of the diffusion process
Facilitates comparison of different conversion conditions
Improves understanding and explainability of singing voice conversion
Abstract
In this study, we present SingVisio, an interactive visual analysis system that aims to explain the diffusion model used in singing voice conversion. SingVisio provides a visual display of the generation process in diffusion models, showcasing the step-by-step denoising of the noisy spectrum and its transformation into a clean spectrum that captures the desired singer's timbre. The system also facilitates side-by-side comparisons of different conditions, such as source content, melody, and target timbre, highlighting the impact of these conditions on the diffusion generation process and resulting conversions. Through comparative and comprehensive evaluations, SingVisio demonstrates its effectiveness in terms of system design, functionality, explainability, and user-friendliness. It offers users of various backgrounds valuable learning experiences and insights into the diffusion model…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMusic and Audio Processing
