Reporting Errors and Common Issues

If you encounter an error while using Applio, this guide will help you report it effectively and find solutions to common problems.

How to Report an Error

Before reporting a new error, please check the Common Errors and Solutions section below to see if your issue is already known.

If you can’t find a solution, you can report the error in two ways:

Method 1: GitHub Issues

This is the preferred method for reporting bugs.

1. Go to the Applio GitHub Issues page.
2. Click on New Issue.
3. Provide a clear and descriptive title for the issue.
4. In the description, include a screenshot of the error and the full text from the Applio command line window.
5. Describe the steps you took to encounter the error.

Method 2: Built-in Bug Reporter

You can also use Applio’s built-in bug reporter.

Note

You can also get help from the community by joining our Discord Server and asking in the help channels.

---

Common Errors and Solutions

Installation and Dependency Errors

Tip

Applio is installed in a self-contained Conda environment with all the necessary dependencies. You shouldn’t need to install anything else manually.

• Miniconda Download Fails: If the automatic download of Miniconda fails, you can download and install it manually.
• Conda “Wheels” Errors: This usually means that the Visual Studio Build Tools are not installed correctly. The Applio installer should handle this automatically, but you may need to install them manually.
• TensorBoard or Dependency Errors: These errors usually indicate that the dependencies were not installed correctly. Make sure you are using the latest version of the installer and that your system meets the installation requirements.

Installation Requirements Checklist

• Path Issues: Did you install Applio in a directory with spaces or special characters in the path (e.g., C:\Program Files\Applio)? If so, please reinstall it in a simple path like C:\Applio.
• Antivirus/Firewall: Did you have your antivirus or firewall enabled during installation? These can sometimes interfere with the installation process. Try reinstalling with them temporarily disabled.
• Administrator Privileges: Did you run the run-applio.bat file with administrator privileges? If so, try running it normally, without administrator rights.

Runtime Errors

• FFmpeg and UTF-8 Errors: These errors are often caused by special characters in file paths or filenames. Make sure all your file and folder names use standard ASCII characters.
• Connection Issues: If you can’t connect to the Applio interface, make sure the console window is running and that your firewall is not blocking the connection.
• Proxy Issues: If you’re using a system-wide proxy, try disabling it and then refreshing the Applio interface.
• GPU Memory Issues: If you run out of GPU memory during training, try reducing the Batch Size. If it happens during inference, you may need to adjust your settings in the config.py file. Also try enabling FP16 precision in the settings tab, which reduces VRAM usage by approximately 40%.
• CPU Processing Issues: If your CPU is struggling, try reducing the Number of CPU threads in the training settings. You can also pre-process your dataset into shorter audio clips.
• Model Not Found in List: If a model you placed in the logs folder doesn’t appear in the UI dropdowns, click the Refresh button next to the model selector. If it still doesn’t appear, check that the .pth file is valid by running python core.py model_information -m path/to/model.pth in the terminal.
• “No module named ‘rvc’” Error: This typically means you’re running core.py or another script from outside the Applio directory. Make sure your terminal is in the Applio root directory before running any commands.
• Index File Loading Error: If the index file fails to load, it may be corrupted or incompatible. Try regenerating it by going to the Extras tab and using the Generate Index feature, or run python core.py index -n your_model_name.
• Audio Artifacts / Robotic Sound: This is usually caused by incorrect settings. Try: using a different pitch extraction algorithm (switch from RMVPE to CREPE or vice versa), lowering the Search Feature Ratio (try 0.5), enabling “Clean Audio,” or ensuring the embedder matches the one used during training.
• Slow Inference: If inference is very slow, try: enabling FP16 precision, using CREPE-tiny instead of full CREPE for pitch extraction, disabling post-processing effects, or reducing the hop length.
• Port Already in Use: If you see an error about port 6969 being in use, it means another instance of Applio (or another application) is running on that port. Close the other instance or change the port using the --port flag: run-applio.bat --port 6970.