Repository navigation
Releases: nsijwali/yamlreel
Release list
yamlreel 0.2.1: now on PyPI
yamlreel is now on PyPI.
pip install "yamlreel[ffmpeg]" # or: pipx install "yamlreel[ffmpeg]"
yamlreel init system-design my-video.yaml
yamlreel render my-video.yaml --preview🎬 The 1-minute demo, made entirely by yamlreel from one YAML file, is on the v0.2.0 release.
What's new
- First PyPI release:
pip install "yamlreel[ffmpeg]"brings its own ffmpeg; add[capture]for web-page screenshots. - Releases are published automatically from GitHub with Trusted Publishing: tested, built, smoke-tested from the wheel, and approved before upload.
- README works on PyPI too, with a short "Why yamlreel" comparison.
- Contributing is easier: CONTRIBUTING.md, a code of conduct, and bug and feature forms.
Upgrading from vidgen? See the v0.2.0 release notes.
yamlreel 0.2.0
Write a short YAML file. Get a finished video. yamlreel renders system-design explainers, code tips and project demos for YouTube, Shorts and LinkedIn: animated diagrams, typed code, original music and an optional voiceover in your own voice, all on your own machine.
🎬 Demo: the attached videos were made entirely by yamlreel from one YAML file, with no video editor. yamlreel-demo-vertical.mp4 is the Shorts cut and yamlreel-demo-landscape.mp4 the YouTube cut.
Highlights
- Renamed from vidgen to yamlreel. "vidgen" clashed with hundreds of other projects; yamlreel is unique.
- Easier install:
pip install -e '.[ffmpeg]'brings its own ffmpeg, so there's nothing else to install on most systems. --no-voicerenders any spec with music only, no text-to-speech server needed.- Clearer errors for a missing ffmpeg, missing fonts or a bad voice-server URL, each telling you how to fix it.
- Per-user font folders are found on macOS and Windows; Fedora and Arch font paths are supported.
Upgrading from vidgen
Old names keep working until 0.3.0, with a warning: the vidgen command, VIDGEN_* variables, ~/.config/vidgen/config.yaml and ./vidgen.config.yaml. To switch over:
git pull
pip uninstall -y vidgen && pip install -e '.[ffmpeg]' # with your venv active
mv ~/.config/vidgen ~/.config/yamlreel # if you have a user configThe old GitHub URL redirects here. Set VIDGEN_* variables as YAMLREEL_*, and rename vidgen.config.yaml to yamlreel.config.yaml in your projects. If VoiceStudio binds a voice to the client vidgen, set voice_server.client_id: vidgen or rebind it.
Get started
git clone https://github.com/nsijwali/yamlreel.git && cd yamlreel
python3 -m venv .venv && source .venv/bin/activate
pip install -e '.[ffmpeg]'
yamlreel init system-design my-video.yaml && yamlreel render my-video.yamlDemo voice: cloned with VoiceStudio (OmniVoice model, CC-BY-NC). Music: generated by yamlreel.