Update README.md

This commit is contained in:
2025-07-16 22:23:14 +02:00
parent 7e64b3a30b
commit 47cf1c6dd5

View File

@@ -0,0 +1,62 @@
# anime_audio_copilot.py
## Overview
`anime_audio_copilot.py` is a batch-processing script for MKV files, designed for anime encoding workflows. It automates video encoding (with AV1 via SVT-AV1 and `av1an`), audio conversion (to Opus with normalization and optional downmixing), and preserves detailed per-file logs. The script also handles VFR-to-CFR conversion using HandBrakeCLI when needed.
## Features
- **Video encoding**: Uses UTVideo intermediate, VapourSynth, and AV1 encoding via `av1an` and SVT-AV1.
- **Audio processing**: Converts non-Opus/AAC tracks to normalized Opus, with optional downmixing for surround audio.
- **VFR handling**: Detects Variable Frame Rate sources and converts to Constant Frame Rate using HandBrakeCLI.
- **Logging**: Creates a log file for each processed MKV in the `conv_logs` directory.
- **Cleanup**: Moves original files to `original/`, completed files to `completed/`, and deletes temporary files.
## Requirements
The following tools must be installed and available in your `PATH`:
- ffmpeg
- ffprobe
- mkvmerge
- mkvpropedit
- sox
- opusenc
- mediainfo
- av1an
- HandBrakeCLI
## Usage
Place your `.mkv` files in the working directory. Run the script:
```sh
python3 anime_audio_copilot.py
```
### Optional Arguments
- `--no-downmix`
Preserve original audio channel layout (do not downmix surround audio to stereo).
Example:
```sh
python3 anime_audio_copilot.py --no-downmix
```
## Output
- Processed files are moved to the `completed/` directory.
- Originals are moved to the `original/` directory.
- Per-file logs are saved in the `conv_logs/` directory.
## Notes
- The script will skip files that cannot be decoded by ffmpeg.
- Temporary files are cleaned up automatically after processing.
- Scene detection and AV1 encoding may take a long time depending on file size and system performance.
## License
See repository for