mikezzb/lyrics-sync

A deep learning lyrics-to-audio alignment system, generating synchronized lyrics from a song and its lyrics

36
/ 100
Emerging

Uses a three-stage pipeline: vocal separation via source separation models, singing voice recognition via transfer learning from speech domain (fine-tunable Wav2Vec2), and forced alignment to produce word-level `.lrc` timestamps. Outputs synchronized lyrics with precise onset/offset timing for each word, enabling applications like karaoke systems and music analysis tools.

No Package No Dependents
Maintenance 10 / 25
Adoption 8 / 25
Maturity 9 / 25
Community 9 / 25

How are scores calculated?

Stars

66

Forks

5

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jan 12, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mikezzb/lyrics-sync"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.