Skip to content
SPECIFICTION
Get listed

Diff-SVC

github.com

WindowsLinuxSelf-hosted
Visit github.com
Spec score 7.4 #34 of 55 in AI song cover generators

Spec sheet

Purpose
Diff-SVC converts input singing voices into a target timbre and supports basic pitch correction.Read 9 Oct 2026 · github.com
Integrations
The maker identifies DiffSinger, the OpenVPI-maintained DiffSinger version, and soft-vc as project foundations.Read 9 Oct 2026 · github.com
Security
The repository page displays no reported security advisories in its Security and quality section.Read 9 Oct 2026 · github.com
Support
The README points users to a QQ channel for project files and an English-language Discord for support.Read 9 Oct 2026 · github.com
License
The repository identifies its license as AGPL-3.0.Read 9 Oct 2026 · github.com
Attribution
The maker asks users who redistribute the code or publicly publish results to credit the original author and repository.Read 9 Oct 2026 · github.com
Audio formats
The training guide says data preparation supports WAV and Ogg audio.Read 9 Oct 2026 · github.com
Audio sample rates
The README records support for 44.1 kHz audio.Read 9 Oct 2026 · github.com
Custom datasets
The project supports training on custom datasets.Read 9 Oct 2026 · github.com
Dependencies
The README credits DiffSinger, the OpenVPI-maintained DiffSinger version, and Soft-VC as project bases.Read 9 Oct 2026 · github.com
Inference
Inference is available through the included inference.ipynb notebook or infer.py script.Read 9 Oct 2026 · github.com
Intended use
The maker says the project is for academic exchange and is not intended for production environments.Read 9 Oct 2026 · github.com
Long audio
Inference automatically slices audio longer than 30 seconds at silences.Read 9 Oct 2026 · github.com
Maintenance
The project is no longer actively maintained, and its README describes it as an early exploration of diffusion models for singing voice conversion.Read 9 Oct 2026 · github.com
Method
The project describes itself as singing voice conversion via diffusion models.Read 9 Oct 2026 · github.com
Performance
The guide says full inference can run on a 6 GB GTX 1060 GPU.Read 9 Oct 2026 · github.com
Real-time conversion
The README mentions support files for real-time voice conversion inference.Read 9 Oct 2026 · github.com
Redistribution
The README asks users who redistribute the code or publicly publish results to credit the original author and repository.Read 9 Oct 2026 · github.com
Training
The README provides a command to train a model using a configuration file and a user-supplied experiment name.Read 9 Oct 2026 · github.com
Training data
The guide supports single-speaker training and recommends at least three hours of target-speaker dry vocals.Read 9 Oct 2026 · github.com

Where Specifiction places it

Alternatives to Diff-SVCAll alternatives →