CJK lyric support
Japanese, Chinese, and Korean lyrics get per-character forced alignment with romanized readings (Hepburn, pinyin, Revised Romanization) shown above each token.
Turn any song into karaoke. Nightingale separates vocals, aligns lyrics, and gives you pitch scoring, key/tempo controls, and rich backgrounds in one self-contained app.
Features
Explore docs🎤 Stem separation
Vocals are isolated from instrumentals using the UVR Karaoke model or Demucs. Guide vocal volume is adjustable.
📝 Word-level lyrics
WhisperX transcribes and aligns every word to the audio. Existing lyrics from LRCLIB are used when available.
🈯 CJK lyrics
Japanese, Chinese, and Korean songs get per-character forced alignment with romanized readings (Hepburn, pinyin, Revised Romanization) shown above each token.
🗣️ Pluggable ASR
Choose Whisper (default, broad coverage) or Parakeet v3 (experimental) for ~25 European languages. NeMo on CUDA, ONNX everywhere else.
🎼 UltraStar songs
Drop USDX song folders into your library — pitch and lyrics come straight from the file, no analyzer pass needed. Experimental.
🎯 Pitch scoring
Sing into your mic and get scored in real-time. Star ratings and per-song scoreboards track your progress.
🎚️ Key and tempo shifts
After analysis, adjust key and tempo from the song list to match your voice and comfort level.
👤 Player profiles
Multiple profiles with separate score histories. Switch between singers without losing anyone's records.
🎬 Video file support
Drop .mp4 or .mkv files into your library. Vocals are separated and the original video plays as the background.
🌌 Audio-reactive backgrounds
10 GPU shaders that react to your mic input in real time, 5 Pixabay video flavors, plus the source video for video files.
🧭 Sidebar filters
Browse by quick filters, metadata buckets, artists, and albums. Analyze All helps queue your library faster.
🎙️ Mic mirroring
Optional live mic monitor during playback for practice sessions, with an adjustable monitor gain.
🎮 Gamepad support
Navigate menus, pick songs, and control playback entirely with a controller. D-pad, sticks, face buttons.
📦 Single binary
ffmpeg, Python, PyTorch, and the ML models are all bootstrapped on first launch. Nothing to install.
🗂️ Flexible data folder
Choose where Nightingale stores cache, models, videos, and library data during setup.
⬆️ In-app updates
On macOS and Windows: auto-checks at launch, badges the sidebar avatar when a release is available, and installs signed updates with one click. On Linux the Update entry opens GitHub Releases for a manual download.
What's new in v0.6
See full docsParakeet v3 ASR (experimental)
Alternative to Whisper for ~25 European languages. NeMo on CUDA, ONNX Runtime on CPU and Apple Silicon. Pick it from Settings.
UltraStar Deluxe songs (experimental)
Drop USDX song folders into your library and play them with their built-in pitch and lyric data. No analyzer pass needed.
Audio-reactive shaders
Ten GPU backgrounds (five new) react to your microphone in real time — louder vocals push the visuals harder.
Built-in updater
On macOS and Windows, Nightingale checks for new releases on launch, badges the sidebar avatar when one is ready, and installs signed updates with one click. Linux ships without the updater — the Update entry opens GitHub Releases for a manual download.
How it works
Read pipeline docsSeparate
UVR Karaoke or Demucs splits the track into vocals and instrumental. Audio is extracted from video files automatically.
Transcribe
Synced lyrics are looked up on LRCLIB first. If nothing's found, WhisperX (or Parakeet v3, experimental) transcribes the vocals with word-level alignment. CJK songs get romanized readings above each token.
Play
Playback includes highlighted lyrics, pitch scoring, key/tempo controls, mic mirroring, and dynamic backgrounds.
Platforms
Runs on Linux, macOS, and Windows. GPU acceleration via CUDA or Metal when available, CPU fallback everywhere else.
Linux x86_64, aarch64
macOS ARM, Intel
Windows x86_64