Turn your music library into a karaoke stage.
An open-source desktop karaoke app powered by on-device AI stem separation and synced lyrics.
I love singing karaoke at home, but every existing solution has its own set of problems.
The most mature option is probably Karafun — a paid service that sidesteps copyright by re-recording famous songs. That's neat, but it comes with issues:
- Their re-recorded instrumentals inevitably sound a little different from the originals
- Their catalog doesn't always include the niche songs I want to sing
- I hate subscriptions
Then there's Apple Music Sing, which offers on-device vocal removal for karaoke. Also neat — but Apple Music is yet another subscription, and I hate subscriptions.
To dodge the subscription trap, you could go the more traditional route — something like OpenKJ for playing CD+G/media+g files. But CD+G files are niche, hard to find, and have to be purchased separately.
That pretty much leaves scouring YouTube for karaoke videos of dubious origin and questionable copyright status. Not exactly a unified experience, and the song I want is missing half the time.
So my no-compromise solution was born: OpenKara uses open-source AI to separate the digital music you already own in unencrypted form — whether it's from CD rips, Bandcamp, Qobuz, iTunes, or your local library's music service. I know there are plenty of people who, like me, prefer to buy once and own forever. OpenKara turns my existing music library into a karaoke library, so I don't have to pay for KTV, and my catalog is shaped by my own taste — not the mainstream.
- Local Audio Import — Use music you already own. No subscriptions, no repurchases.
- AI Stem Separation — Separate vocals and accompaniment on-device.
- Synced Lyrics — Load timed lyrics from online sources, embedded tags, or sidecar
.lrcfiles. - CD+G Sidecars — Render same-name
.cdggraphics during fullscreen playback when a track includes them. - Portable Library — Self-contained library directory that works on NAS, USB drives, and across machines.
- Cross-Platform — Available on macOS, Windows, and Linux.
- 4-Stem Mixer — Individual volume control for vocals, drums, bass, and other instruments. Collapsible accompaniment slider with per-stem breakdown.
- Dual Separation Modes — Choose between 2-stem (vocals + accompaniment) or 4-stem (vocals + drums + bass + other). Upgrade individual songs from 2-stem to 4-stem on demand.
- Efficient Stem Storage — Separated stems are cached compactly to keep library storage practical.
- Resumable Separation — Per-chunk checkpointing means separation resumes from where it left off if the app is closed mid-process.
Download the latest build for your platform from GitHub Releases:
| Platform | Format |
|---|---|
| macOS (Apple Silicon) | .dmg |
| macOS (Intel) | .dmg |
| Windows | .exe (NSIS installer) |
| Linux | .AppImage / .deb |
macOS (Homebrew):
brew install thedavidweng/tap/openkaramacOS Gatekeeper note: If macOS says the app is damaged or can't be opened, run:
xattr -rd com.apple.quarantine /Applications/OpenKara.appOn first launch, OpenKara will prompt you to create a Karaoke Library and start downloading the default AI model in the background.
Prerequisites:
git clone https://github.com/thedavidweng/OpenKara.git
cd OpenKara
pnpm install
./scripts/setup.sh # downloads Demucs ONNX model for local dev
pnpm tauri devTo bundle the official Google Drive Desktop OAuth client into a release build,
set OPENKARA_GOOGLE_DRIVE_OAUTH_CLIENT_JSON or
OPENKARA_GOOGLE_DRIVE_OAUTH_CLIENT_JSON_PATH before pnpm tauri build.
OpenKara treats the per-user refresh/access tokens as secrets and stores them in
the OS credential store; for desktop apps, the OAuth client_secret itself
cannot be treated as confidential.
- Source icon:
src-tauri/icons/app-icon.png(1024x1024master asset) - Regenerate all platform icons with
pnpm icons:generate - Generated assets are written to
src-tauri/icons/for Tauri desktop and future mobile targets
OpenKara uses custom ONNX builds of Demucs models for stem separation. Models are maintained in a separate repository:
openkara-models — Reproducible ONNX model conversion pipeline
| Model | Description | Input | Output | Format |
|---|---|---|---|---|
htdemucs |
Standard — Hybrid Transformer Demucs | Stereo audio at 44.1 kHz (7.8s) | 4 stems: drums, bass, other, vocals | ONNX (opset 17) |
htdemucs_ft |
High Quality — Fine-tuned 4-model ensemble | Stereo audio at 44.1 kHz (7.8s) | 4 stems: drums, bass, other, vocals | ONNX (opset 17) |
On first launch, OpenKara automatically downloads the standard openkara-models v2.0.1 asset into the app data directory. The current standard model is ~339 MiB on disk, and the optional high quality model is ~1.32 GiB. Both assets are ONNX Runtime-optimized and carry metadata used for cache invalidation. See the openkara-models README for details on the conversion pipeline. For local development and deterministic tests, run ./scripts/setup.sh to populate src-tauri/models/.
| Layer | Technology | Purpose |
|---|---|---|
| Desktop framework | Tauri 2 | Rust backend + system WebView |
| Frontend | React 19 + TypeScript 5 | UI components |
| Bundler | Vite 7 | Dev server and build |
| Styling | Tailwind CSS 4 | Utility-first CSS |
| State | Zustand | Lightweight global state |
| Audio decode | symphonia | Pure-Rust codec support |
| Audio output | cpal | Cross-platform audio playback |
| AI inference | ONNX Runtime via ort | Demucs v4 stem separation |
| Lyrics | LRCLIB | Open synced lyrics API |
| Metadata | lofty | ID3v2, Vorbis, FLAC tag reading |
| Audio encode | vorbis_rs | OGG/Vorbis stem compression |
| Database | SQLite via rusqlite | Song, lyrics, and stems cache |
┌─────────────────────────────────────────────────────────────┐
│ Tauri Frontend (React) │
│ ┌────────────────┐ ┌──────────────────────┐ │
│ │ File Import │ │ Karaoke Player │ │
│ │ & Library │ │ / Mixer │ │
│ │ │ │ (lyrics sync) │ │
│ └────────────────┘ └──────────────────────┘ │
│ ┌────────────────┐ ┌──────────────────────┐ │
│ │ Playback │ │ Progress & Volume │ │
│ │ Controls │ │ Controls │ │
│ └────────────────┘ └──────────────────────┘ │
├─────────────────────────────────────────────────────────────┤
│ Tauri Rust Backend │
│ ┌────────────────┐ ┌──────────────────────┐ │
│ │ Audio Decode │ │ AI Stem Separation │ │
│ │ & Playback │ │ (Demucs v4 / ONNX) │ │
│ └────────────────┘ └──────────────────────┘ │
│ ┌────────────────┐ ┌──────────────────────┐ │
│ │ Metadata │ │ Lyrics Fetcher │ │
│ │ Reader │ │ (LRCLIB + embedded) │ │
│ └────────────────┘ └──────────────────────┘ │
│ ┌───────────────────────────────────────────────────────┐ │
│ │ Portable Library │ │
│ │ (SQLite + media files + stems) │ │
│ └───────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
| Format | Import | CD+G Graphics | Stem Separation |
|---|---|---|---|
| MP3 | ✅ | Same-name .cdg sidecar |
✅ |
| FLAC | ✅ | Same-name .cdg sidecar |
✅ |
| WAV | ✅ | Same-name .cdg sidecar |
✅ |
| OGG / Vorbis | ✅ | Same-name .cdg sidecar |
✅ |
| AAC / M4A | ✅ | Same-name .cdg sidecar |
✅ |
| MP3+G ZIP | ✅ | Embedded audio + .cdg bundle |
N/A |
OpenKara imports same-name audio + .cdg pairs as managed CD+G tracks, and it can also import MP3+G ZIP archives directly. Standard audio tracks are resampled to 44.1 kHz stereo for the Demucs model. Managed CD+G tracks already contain accompaniment-only audio, so they skip stem separation.
OpenKara stores all data in a self-contained library directory:
MyKaraokeLibrary/
├── .openkara-library # marker file
├── openkara.db # SQLite database
├── media/ # imported standard audio copies
│ └── {hash}.mp3
├── media-g/ # managed CD+G assets
│ ├── {hash}.mp3 # paired audio for CD+G playback
│ ├── {hash}.cdg # paired CD+G graphics sidecar
│ └── {hash}.zip # MP3+G ZIP archive when imported as a bundle
└── stems/ # separated tracks for standard audio imports
└── {hash}/
├── vocals.ogg
├── accompaniment.ogg # 2-stem mode
├── drums.ogg # 4-stem mode
├── bass.ogg # 4-stem mode
└── other.ogg # 4-stem mode
All paths in the database are relative — including CD+G sidecars and MP3+G ZIP assets — so the whole library can be moved to a NAS, USB drive, or network share and opened by any OpenKara instance on any OS. Per-machine configuration (library location) is stored separately in the app data directory.
- Project scaffolding (Tauri 2 + React + TypeScript + Vite)
- SQLite database with migration system
- Audio import with metadata extraction (ID3v2, Vorbis, FLAC)
- Library search and browsing
- Audio decode and playback (symphonia + cpal)
- Playback state machine (play / pause / seek / volume)
- Demucs v4 ONNX stem separation with progress tracking
- Stems caching (hash-based, no re-inference on replay)
- Karaoke mode toggle (original / instrumental)
- Synced lyrics fetch (LRCLIB → embedded → sidecar .lrc)
- Lyrics display with rAF-based sync and click-to-seek
- Per-song lyrics timing offset
- First-launch AI model bootstrap with background download
- Portable library system with relative paths
- Full frontend UI (sidebar, player, lyrics panel, settings)
- Queue panel with play next, drag reorder, and auto-advance
- Keyboard shortcuts (space, arrows)
- Drag-and-drop file import
- CI/CD pipeline (macOS, Windows, Linux)
- Release automation (tag → GitHub Release with binaries)
OpenKara v0.2.0 is the release that established the current core app flow.
- CD+G sidecar playback for same-name audio +
.cdgpairs - MP3+G ZIP import and playback support
- Managed CD+G library storage and pairing disambiguation
- Second-display fullscreen audience window
- 4-stem volume mixer with collapsible UI
- Dual separation modes (2-stem / 4-stem) with settings persistence
- Efficient compressed stem storage
- Resumable separation with per-chunk checkpointing
- Multi-threaded ONNX inference optimization
- Settings system (stem mode configuration)
- UI polish and transitions
- Error toasts and user-facing error messages
- App icon and branding
OpenKara v0.4.0 is the current stable release. It adds:
- AirPlay support for casting playback to compatible devices
- Improved player behavior and layout at narrow window widths
- Visual refinements to the Windows app appearance
- Better preservation of original track metadata on import
- WinGet installation support on Windows
- Mic Input & Vocal Effects — Microphone capture, reverb, echo, volume mix
- Saved Playlists & Singer Rotation — Named playlists, singer assignment, and stronger turn-based queue workflows
- Pitch & Key Shift — Real-time pitch shifting of the accompaniment track
- Session Recording — Record vocal performances, export as audio
- CJK Transliteration — Romaji / Pinyin display alongside original lyrics
- Node.js 20+
- pnpm 10+
- Rust stable via rustup
- Tauri 2 prerequisites for your platform
pnpm install
./scripts/setup.sh # download Demucs ONNX model to src-tauri/models/
pnpm tauri dev # start dev server with hot reloadscripts/setup.sh places the model in src-tauri/models/ for local development and deterministic testing only. End-user installs use the app data directory for runtime model downloads.
cd src-tauri && cargo test # backend tests (70+ tests)
pnpm lint # ESLint
pnpm format # Prettier checkpnpm tauri build # production build with platform-specific bundle- Pushes to
maintrigger the CI workflow (.github/workflows/ci.yml) — lint, build, and test on macOS, Windows, and Linux. - Pushing a version tag (e.g.
v0.4.0) triggers the release workflow (.github/workflows/release.yml) — builds and attaches binaries to a GitHub Release.
- Docs Hub — Canonical index for design docs, plans, specs, references, and site content
- Architecture — System design, tech stack, data flow, and runtime details
- Project Structure — Current directory layout and module responsibilities
- Development Phases — Phase checklist with verification steps
- Technical Roadmap — Technology choices, API contracts, and risk mitigations
- Milestones — Milestone task table with exit criteria
Contributions are welcome! Please open an issue before starting major changes so we can discuss the approach.
- Fork the repository
- Create a feature branch (
git checkout -b feature/my-feature) - Make your changes and ensure tests pass (
cargo test) - Submit a pull request
- Demucs — AI stem separation model by Meta Research
- openkara-models — ONNX model conversion pipeline for OpenKara
- demucs.onnx — Reference for STFT/ISTFT to real-valued ONNX conversion
- LRCLIB — Open synced lyrics API
- monochrome — Lyrics sync and LRCLIB integration reference
MIT — Copyright (c) 2025 David Weng