guxm2021/ALT_SpeechBrain

[ISMIR 2022] Transfer Learning of wav2vec 2.0 for Automatic Lyric Transcription

36
/ 100
Emerging

Adapts wav2vec 2.0 self-supervised representations to singing through transfer learning, exploring optimal layer-wise fine-tuning strategies to bridge the ASR-to-ALT domain gap. Extends the standard CTC decoder with a hybrid CTC/attention architecture to improve character-level transcription accuracy on singing audio. Built on the SpeechBrain toolkit with PyTorch, supporting training on multiple benchmark datasets (DSing, DALI, Hansen, Jamendo, Mauch) including source-separated vocal tracks via Demucs.

No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

49

Forks

6

Language

Python

License

Apache-2.0

Last pushed

May 07, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/guxm2021/ALT_SpeechBrain"

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