Self-host Nightingale
v1.3.0Karaoke app with AI stem separation and pitch scoring
- Stars
- 1 521
- Port
- 8080 (HTTP server inside container)
razzaru/nightingale
What is Nightingale?
Nightingale is a karaoke application that transforms your personal music library into a karaoke experience using machine learning. It separates vocals from instrumentals using UVR Karaoke or Demucs models, finds or transcribes synchronized lyrics via LRCLIB or WhisperX, and provides real-time pitch scoring with star ratings. The app supports local multiplayer for up to four singers with separate microphones, adjustable key and tempo, customizable backgrounds including GPU shaders and Pixabay videos, and player profiles with score tracking. It runs as a native desktop application on Linux, macOS, and Windows, or as a self-hosted web server on Linux accessible from any device on the LAN. The self-hosted mode uses a single binary that downloads its own isolated Python environment and ML models on first launch. No external database or cache services are required — it uses embedded SQLite for the library and caches analysis artifacts locally.
Should you self-host Nightingale?
Nightingale scores 58/100 on selfhost.directory (how scoring works) and ranks #15 of 54 in Media Streaming. It has 1 521 GitHub stars.
It shipped 9 stable releases in the past year. The latest, v1.3.0, came out on 15 October 2026. New versions land roughly every 3 weeks.
Setup is advanced: plan for several services and some ongoing maintenance. An official Docker image is available. It is GPL-3.0-licensed and written mostly in Typescript.
Its closest alternative on selfhost.directory is Invidious (81/100, 25 799 stars). It scores higher overall, so compare the two before you pick. See Nightingale vs Invidious side by side.
Features
- CJK lyrics with romanization
- Customizable backgrounds
- Gamepad and touch support
- Key and tempo adjustment
- Local multiplayer
- Microphone monitoring and latency calibration
- Player profiles
- Real-time pitch scoring
- Self-hosted web mode
- Session mode queue management
- Shared analysis cache
- Synchronized lyrics
- UltraStar Deluxe importExperimental
- Vocal Separation
Nightingale README
More in Audio Processing
OmniVoice-StudioThis self-hosted, open-source ElevenLabs alternative clones and designs voices, dubs video and builds audiobooks in 646 languages, fully local on desktop or Docker.
OpenBriefFree open-source desktop app for macOS, Windows, Linux that imports video/audio, extracts transcripts locally, generates grounded summaries, and lets you chat.
openclaw-voiceSelf-hosted browser voice interface for AI assistants. Runs Whisper locally, uses ElevenLabs or local TTS, connects to OpenAI, Claude, or custom agents.
OpenTranscribeOpenTranscribe is a self-hosted, open-source AI transcription platform with speaker identification, cross-video matching, hybrid search, and AI summarization.
Retrieval-based-Voice-Conversion-WebUISelf-hosted, open-source voice timbre conversion framework that trains a VC model from around 10 minutes of audio and offers a WebUI plus a real-time voice-changing GUI.
ScribeCrateAn open-source transcription server you run yourself, offering OpenAI-compatible speech-to-text, translation, and subtitle generation via Docker on CPU or GPU.
ScriberrSelf-hosted audio transcription with NVIDIA Parakeet, Canary, Whisper. Speaker diarization, AI chat via Ollama/OpenAI, YouTube transcription, PWA, dark mode.
StemDeckOpen-source stem separator and practice studio for musicians: split a song into six parts locally with Demucs, then mute, slow, loop and transpose it.
subarrSelf-hosted open-source service coordinating subtitle coverage, audio-language verification, and Whisper transcription across Sonarr, Radarr, Bazarr, and subgen.
SubgenSelf-hosted subtitle generator using OpenAI Whisper. Integrates with Bazarr, Plex, Jellyfin, Emby, Tautulli. Runs on CPU or NVIDIA GPU via Docker.
tele-vdo-riderSelf-hosted Telegram bot that converts videos from YouTube and other sites to MP3 using youtube-dl and ffmpeg, with automatic file splitting for Telegram's size limits.
transcriberSelf-hosted, open-source web app for transcribing audio/video with Whisper. Supports YouTube, local files, real-time progress, dark mode.
0 comments