[Feature]: How to convert a PTH denoising model to ONNX?

[jackwxs]

Description

How to convert a PTH denoising model to ONNX?

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

• Create Plan

Examples

• Example 1
• Example 2

---

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord or schedule a call!

[beveradb]

Why? 😅 and which model?

There are various model architectures supported by audio-separator which use models in several different formats.

The packaging format is less relevant really, it's the inference code which needs to work with whatever model you're trying to use.

[jackwxs]

为什么？😅 还有，是哪个型号？

它支持多种模型架构，audio-separator这些架构使用几种不同格式的模型。

打包格式其实不太重要，重要的是推理代码需要与你尝试使用的任何模型兼容。

I need to use C++ for inference in UVR-DeNoise-Lite.pth, should I switch to ONNX?

[jackwxs]

However, I cannot find the structure of the UVR-DeNoise-Lite.pth model definition and cannot convert it.