Skip to content

API reference

syncalong exposes a small, curated public API. Everything below is re-exported from the package root, so you can write syncalong.align(...) rather than reaching into submodules.

The package ships a py.typed marker, so type checkers (mypy, pyright) resolve these signatures in your own code.

Where to start

If you want to… Use
Align one song and get a structured result align()AlignmentResult
Get just the LRC string align_to_lrc()
Load the Whisper model once and reuse it Transcriber
Parse a lyrics file/text yourself parse_lyrics() / parse_lyrics_text()
Run the aligner on your own transcript align_lyrics_to_transcript()
Render timed lines to LRC format_lrc()

Public names

The complete public surface (syncalong.__all__):

  • High-level facade: align, align_to_lrc, AlignmentResult
  • Transcription: Transcriber, transcribe_audio, WordTimestamp
  • Lyrics: parse_lyrics, parse_lyrics_text, lyrics_prompt, LyricLine
  • Low-level building blocks: align_lyrics_to_transcript, format_lrc, separate

Use the navigation to browse each module in detail.