Repository navigation
Install Windows
-
Tools
winget install astral-sh.uv winget install Gyan.FFmpeg # the "full" build: libass captions, loudnorm, QSV/NVENC
Open a new terminal so both are on PATH.
-
Code: copy the repo folder (everything except
streams/,.venv/,logs/) toC:\clipper(or anywhere), then in it:uv sync --extra gpu --extra faces --extra discord # worker + serve + bot; a bot-only machine needs only --extra discord
The
gpuextra pulls torch 2.8 with CUDA 12.8 from the PyTorch index (pinned in pyproject.toml). The CUDA toolkit is not needed; the wheels bring the runtime and cuDNN. Ifclippa doctorsays torch sees no CUDA GPU, the CPU wheel slipped in:uv pip install --reinstall torch==2.8.0 torchaudio==2.8.0 torchvision==0.23.0 --index-url https://download.pytorch.org/whl/cu128
-
Config: copy
clipper.tomlfrom the main machine. Things to change per machine:-
[paths] streams: the NAS share, e.g."//nas/clipper/streams"or a mapped drive. Every machine must see the same streams folder. -
[cut] clip_codec:h264_qsvneeds an Intel GPU; on an NVIDIA-only desktop useh264_nvenc(driver 610+ for ffmpeg 8.1) orlibx264.clippa doctortest-encodes it. -
[transcribe] compute_type = "int8_float16"andbatch_size = 4on cards under 8 GB. Anything can also be set per machine inclipper.local.toml(whatclippa setupwrites) or with an env varCLIPPER_<SECTION>_<KEY>, e.g.CLIPPER_CUT_CLIP_CODEC=libx264, so one shared clipper.toml can stay identical.
-
-
Secrets: copy
secrets/(discord_token.txt,youtube_client_secret.json,youtube_token.json; only the machine that uploads needs the YouTube files). -
Keys as user environment variables (System > About > Advanced system settings > Environment Variables, or):
[Environment]::SetEnvironmentVariable('ANTHROPIC_API_KEY', '<key>', 'User') [Environment]::SetEnvironmentVariable('KITSU_TOKEN', '<Kitsu bot token>', 'User')
Add
ANTHROPIC_WORKSPACE_IDif the key isn't scoped to a workspace. Open a new terminal. -
Fonts: thumbnails and Short captions use Impact and Bahnschrift, both standard on Windows. On Linux install
ttf-mscorefonts-installer(Impact) and Roboto Condensed (fonts-roboto-unhinted). The weekly issue needs a few more.Optional: cover layouts and photos, the cast library, faces.
-
Check:
clippa doctor --role gpu(or--role serve,--role bot). -
Run unattended. Serve and the bot run as containers (Docker compose,
deploy/hub, on the NAS or any Docker host: Roles and services). On Windows the scheduled task is for the GPU worker (at logon, restarted if it exits, no admin needed; the scripts sharedeploy\common.ps1, which lists the user env vars the task copies in, since scheduled tasks don't always see ones set after logon: a new key goes there too):deploy\worker\worker-task.ps1 -Install # GPU worker
-Status,-Restart,-Stop,-Uninstallwork. Logs go tologs\.clippa servicesshows the task plus the recorder and any NAS cuts, with serve's and the bot's heartbeats from the share;clippa services restart workerruns the same script, andserveorbotover ssh once[serve] hubnames the compose folder. The bot reaches serve at[serve] url(empty = this machine); on different machines set it, give both the same[warroom] token(orsecrets/warroom_token.txt), and put[paths] scheduleand[paths] newsunder the streams share on both (the bot reads what serve writes there;clippa doctor --role botchecks).
Home · Getting started · clippa doctor · These pages are built from docs/wiki/ in the repo, so edits made here are overwritten: how to edit them
Run it
- Overview
- Getting started
- Roles and services
- Configuration
- Pipeline
- Review
- Reviewing
- War room
- Clippa messenger
- Scheduling and uploads
- Costs
- Checkup
- Audit
- Backups
Install
- Install
- Install checklist
- Kitsu
- Discord app
- YouTube sign-in
- Windows
- macOS
- Transcribe on Modal
- NAS recorder
- NAS hub
- Claude on the NAS
- Phone access
- Model providers
- Adding a show
- YouTube channel
Features
- Pre-shows
- Fetching VODs
- Speakers
- Faces
- Covers
- Cast library
- B-roll
- Clip art
- Weekly issue
- X posts
- TikTok
- News watcher
- Character pieces
- Documentaries
Research
Working on Clippa