Enhance audio verification features and improve code structure

- Added `ffmpeg` as a dependency in `flake.nix` and `package-cleanup.nix` for audio file property analysis.
- Updated `main.py` to include new functions for retrieving audio file properties using `ffprobe` and verifying audio matches with detailed confidence scoring.
- Refactored fingerprint comparison logic to improve accuracy and added logging for better traceability.
- Enhanced the `find_duplicate_singles` function to support audio verification results and confidence scores, providing clearer output for users.
This commit is contained in:
Danilo Reyes
2025-11-13 23:21:12 -06:00
parent 026c7fe0d8
commit 03e8eb6f4e
3 changed files with 450 additions and 120 deletions

View File

@@ -12,9 +12,17 @@ pkgs.python3Packages.buildPythonApplication {
requests
python-dotenv
];
# Runtime dependencies for audio verification
buildInputs = [
pkgs.chromaprint
pkgs.ffmpeg
];
makeWrapperArgs = [
"--prefix PATH : ${pkgs.lib.makeBinPath [ pkgs.ffmpeg pkgs.chromaprint ]}"
];
meta = {
mainProgram = "lidarr-cleanup-singles";
description = "Identify duplicate single tracks in Lidarr";