NOTICE: This is a sample application created with the assistance of my opinionated .claude stack. It's not meant to be used in production, operated for profit, etc. See LICENSE.md for more information.
A web-based audio/video processing tool with 13 filter categories and YAML-driven presets.
- 13 filter categories with 66 presets across audio and video
- Audio filters: Volume, Tunnel (echo), Frequency, Speed, Pitch, Noise Reduction, Compressor
- Video filters: Brightness, Contrast, Saturation, Blur, Sharpen, Transform
- YAML-driven presets via
presets.yml- customize without code changes - Per-file metadata storing settings in
.data/input/{filename}.yml - Dual-handle clip range slider with millisecond precision
- Draggable video preview modal for video files
- URL download via yt-dlp (YouTube, etc.)
- 10 dark themes including Nord, Tokyo Night, Catppuccin, and more
- Configurable via
config.yml
Docker (recommended):
docker compose up -d
# Open http://localhost:8000Or local install:
- Python 3.12+
- ffmpeg (must be in PATH)
- yt-dlp (for URL downloads)
- uv (Astral package manager)
git clone git@github.com:brianclaridge/FFmpeg-sandbox.git
cd FFmpeg-sandbox
uv sync
uv run python -m app.main- Add source files via upload, drag-and-drop, or URL download
- Select clip range using the dual-handle slider
- Choose filter presets from Audio/Video tabs
- Click Process to generate output
- Preview and download the result
| Category | Presets | Description |
|---|---|---|
| Volume | None, 0.25x-4x | Gain control (mute to 4x boost) |
| Tunnel | None, Subtle, Medium, Heavy, Extreme | Echo/reverb effects |
| Frequency | Flat, Bass Cut, Treble Cut, Narrow, Voice | High/low-pass filtering |
| Speed | 0.5x-2x | Playback speed adjustment |
| Pitch | Low, Normal, High, Chipmunk | Pitch shifting (semitones) |
| Noise Reduction | None, Light, Medium, Heavy | Background noise removal |
| Compressor | None, Light, Podcast, Broadcast | Dynamic range control |
| Category | Presets | Description |
|---|---|---|
| Brightness | Dark, Normal, Bright | Image brightness |
| Contrast | Low, Normal, High | Image contrast |
| Saturation | Grayscale, Muted, Normal, Vivid | Color saturation |
| Blur | None, Light, Medium, Heavy | Gaussian blur |
| Sharpen | None, Light, Medium, Strong | Edge enhancement |
| Transform | Flip H/V, Rotate 90/180/270 | Geometric transforms |
Edit presets.yml to add or modify presets:
audio:
volume:
custom:
name: "Custom"
description: "My custom preset"
volume: 2.5Restart container after changes. Presets are validated on startup.
10 dark themes: Darcula (default), Dracula, Monokai, Nord, Solarized, Gruvbox, One Dark, Tokyo Night, Catppuccin, Neon
Theme selection saved in localStorage.
Root Files:
├── config.yml # App configuration
├── presets.yml # Filter presets (66 presets, 13 categories)
├── Dockerfile # Container build
├── docker-compose.yml # Container orchestration
├── Taskfile.yml # Task runner commands
└── pyproject.toml # Python dependencies (uv)
.data/
├── input/ # Source files + per-file metadata
├── output/ # Processed files
└── logs/ # App logs
app/
├── main.py # FastAPI entry point
├── config.py # Config loader
├── models.py # Pydantic schemas (240 lines)
├── routers/
│ ├── audio.py # Processing endpoints
│ ├── download.py # yt-dlp endpoints
│ └── history.py # History endpoints
├── services/
│ ├── presets.py # YAML preset loader
│ ├── processor.py # FFmpeg orchestration
│ ├── metadata.py # File introspection
│ ├── filters_audio.py # Audio filter builders
│ ├── filters_video.py # Video filter builders
│ ├── filter_chain.py # Filter aggregation
│ ├── ffmpeg_executor.py # Subprocess wrapper
│ ├── downloader.py # yt-dlp downloading
│ ├── file_metadata.py # Per-file metadata
│ ├── history.py # Processing history
│ └── settings.py # User settings
├── templates/
│ ├── base.html # Layout + themes
│ ├── index.html # Main interface
│ └── partials/ # 10 HTMX partials
└── static/css/ # 10 modular CSS files
See LICENSE.md
