Browse / 🎬 Creative & Media

stemkit preview

danielravina

stemkit

Split any YouTube song into stems

Stars
1.3k
Language
TypeScript
Category
🎬 Creative & Media

StemKit

Split any YouTube song into isolated stems β€” vocals, drums, bass, guitar, piano and more β€” right on your machine.

Search YouTube or paste a link, pick the instruments you want, and play the result like a mini DAW: the video on one side, every stem on its own fader, all perfectly in sync. Karaoke, acapellas and instrumentals are one click away.

Everything runs locally β€” no accounts, no API keys. Your songs, searches and audio never leave your machine. The app sends a single tiny anonymous ping per install (a random install id + version/OS) so I can count how many people use it β€” see Privacy for details.

platform local

StemKit splitting Queen's Bohemian Rhapsody into six stems β€” video player, presets and color-coded waveform lanes

Features

  • Built-in YouTube search, or paste a link
  • Choose your instruments individually; the right separation engine is picked for you
  • Tight audio/video sync with instant, artifact-free seeking
  • One-click presets: All Β· Karaoke Β· Acapella Β· Drums + Bass
  • Per-stem mute/solo/volume, waveforms with click-to-seek
  • Parallel background splitting with live progress
  • Export checked stems mixed into one file or separately, with mixer volume/pan and original or shifted pitch (WAV, or AAC in Settings)
  • Detected song key (editable) and live pitch shift (Β±6 semitones)
  • Fully offline after setup β€” separation runs on Mac (MPS), NVIDIA GPUs (CUDA), AMD GPUs on Linux (ROCm, experimental) or CPU; ffmpeg included

Download

Grab installers from Releases:

  • macOS: StemKit-x.y.z-mac-arm64.dmg (Apple Silicon) or StemKit-x.y.z-mac-x64.dmg (Intel)
  • Windows: StemKit-Setup-x.y.z.exe (installer) or portable .zip
  • Linux (x64): StemKit-x.y.z-linux-x86_64.AppImage (portable, self-updating) or StemKit-x.y.z-linux-amd64.deb

First launch creates a private Python environment and downloads the separation engine (~2 GB) β€” one time. ffmpeg is bundled β€” nothing else to install.

Optional quality upgrades live behind a gear icon in the app (Settings), each with its own one-time download:

  • Studio-quality vocals (Mel-Band Roformer): +913 MB β€” runs on GPU or CPU (CPU is slower)
  • Fine-tuned demucs (htdemucs_ft): +~320 MB, up to 4Γ— slower
  • Refinement passes: 2 shifts instead of 1, up to 3Γ— slower

macOS first launch: builds are signed with a Developer ID but not notarized, so macOS may say it "cannot verify the developer". One-time fix: System Settings β†’ Privacy & Security β†’ Open Anyway (or xattr -cr /Applications/StemKit.app).

Windows: SmartScreen may warn on first run β€” "More info β†’ Run anyway".

Requirements

  • macOS 12+ (Apple Silicon or Intel x86_64) or Windows 10/11 (x64) or Linux x64 (Ubuntu 22.04+ or equivalent; NVIDIA driver for GPU splits, AMD via ROCm on Linux β€” experimental)
  • No manual installs: if no Python 3.9+ is detected, StemKit downloads a private runtime (python-build-standalone) during first-launch setup
  • Node.js 24+ only for building from source

Intel Mac: PyTorch publishes no macOS x86_64 wheels past 2.2.2, so first-launch setup installs that build instead of the 2.5.1 used everywhere else. MPS is still available on that build when the machine has a Metal-capable GPU (verified on an Intel Mac with a discrete GPU); integrated-only Macs fall back to CPU.

Develop

npm install
npm run dev

Wrong Node version? Scripts auto-relaunch with a suitable one (nvm / nvm-windows).

Build & release

bash scripts/fetch-ffmpeg.sh        # mac (compiles arm64 + x86_64 ffmpeg + libsoxr, needs cmake) / linux (one time)
powershell scripts/fetch-ffmpeg.ps1 # windows (one time)

npm run dist        # mac dmg -> release/
npm run dist:win    # windows nsis+zip -> release/
npm run dist:linux  # linux AppImage+deb (x64) -> release/
npm run dist:all    # both (on the matching OS)

Linux: building the .deb needs dpkg + fakeroot on the host; running the .AppImage needs FUSE. In-app self-update works on the AppImage β€” .deb installs update by re-downloading.

Releases are built by GitHub Actions:

  • push a tag v* β†’ binaries attach to a draft GitHub Release
  • workflow_dispatch ("Run workflow") β†’ on-demand artifacts on the run page

macOS builds are Developer-ID-signed when the certificate is available β€” see Signing in CI below for the one-time setup, plus optional notarization.

Signing in CI (one-time setup)

Local builds sign with your keychain cert automatically. CI runners have empty keychains, so hand them the certificate via repo secrets:

  1. Keychain Access β†’ My Certificates β†’ right-click Developer ID Application: ... β†’ Export β†’ .p12 (set an export password)
  2. Base64 it and add these repo secrets:
    • CSC_MAC_P12 β€” the base64 string: base64 -i developer-id.p12 | pbcopy
    • CSC_MAC_PASSWORD β€” the export password from step 1
  3. Optional (full notarization, zero Gatekeeper prompts): add APPLE_ID, APPLE_APP_SPECIFIC_PASSWORD, APPLE_TEAM_ID and set repo variable ENABLE_NOTARIZATION to true (Settings β†’ Secrets and variables β†’ Actions β†’ Variables). Requires an active Apple Developer membership β€” Apple's notary service rejects expired accounts.

Without these secrets CI falls back to ad-hoc signing (app runs, but Gatekeeper complains on download).

How it works

YouTube URL ──► yt-dlp (+JS runtime) ──► bundled ffmpeg ──► mel-band roformer (vocals) ─┐
                                        β”‚                                               β”œβ”€β–Ί stems/*.wav
                                        └─────────────► demucs htdemucs β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                             (drums/bass/other, shift-averaged)

Electron renderer ◄──── IPC events β”€β”€β”€β”€β”€β”˜
video iframe (muted) + Web Audio stem playback Β· master clock = the audio itself

Notes

  • Downloading audio from YouTube violates their ToS for public products β€” keep this personal.
  • Want the data somewhere else (venv, models, songs, library)? Start the app with STEMKIT_DATA_DIR=/path/to/folder set. Unset keeps the default app folder.
  • yt-dlp breaks occasionally when YouTube changes things; the error dialog offers a one-click update (updates yt-dlp + the challenge solver together).

Privacy

  • Never leaves your machine: the videos you download, the songs you split, your library, your searches and your audio β€” none of it is uploaded anywhere. All separation runs locally.
  • Anonymous install count: the first time the app runs it sends one small POST to a Cloudflare Worker (stemkit-stats.danielravina.workers.dev) β€” once per install, ever. It contains a random install id (generated locally, stored in the app's settings folder), the app version, OS and architecture. No IP-address-based profiles are built, no cookies, no identifiers tied to you, no analytics SDKs. If the first attempt fails (e.g. you're offline), it simply retries on a later launch.
  • What it's for: counting installs and active usage (the same stats you'd get from GitHub release downloads, minus auto-update noise). The counter code is in telemetry-worker/ β€” inspectable like the rest of the app.
  • Fully offline builds: if you'd rather send nothing, build from source and remove src/main/telemetry.ts (or block the worker domain in your firewall) β€” everything else works identically offline.

Layout

src/main         Electron main process (pipeline, env bootstrap, library)
src/preload      IPC bridge
src/renderer     React UI (player, sync engine, waveforms)
python/          separate.py (demucs) and roformer.py (neural vocals) with JSON progress output
python/vendor/   patched model code β€” see python/vendor/README.md
scripts/         node runner, ffmpeg fetchers
build/           icon sources

More in 🎬 Creative & Media

10.8k

JuneYaooo

rea-video

Reverse-engineer any video, shot by shot: filmed, AI-generated or coded, and how to remake it. Breakdowns in the open VIDEO.md format. ι€†ε‘δΈ€εˆ‡θ§†ι’‘οΌŒι€δΈͺι•œε€΄ηœ‹ζ˜―ζ‹ηš„γ€AI η”Ÿζˆηš„θΏ˜ζ˜―δ»£η εšηš„οΌŒε†ε‘Šθ―‰δ½ ζ€ŽδΉˆε€εˆ»οΌ›ζ‹†θ§£η»“ζžœζŒ‰ VIDEO.md 格式输出。