Metraipolis › archive › dc_043ccb74450d

Automatic Lyric Transcription for Greek Songs: Scaling and Task Composition Effects in Whisper Adaptation — arXiv:2609.11302

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11302
License (as archived)
CC0-1.0
Archived
2026-10-16T05:44:07.319Z
SHA-256 of the archived text
0677efd401e73ee174bb2e30d650e8cf8b0f37fd7a3a2216245dbe0ecb0018fe
Stable URL of this page
https://metraipolis.com/documento/dc_043ccb74450d

How to cite

Automatic Lyric Transcription for Greek Songs: Scaling and Task Composition Effects in Whisper Adaptation — arXiv:2609.11302 — arXiv (en), https://arxiv.org/abs/2609.11302 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_043ccb74450d (Metraipolis archive dc_043ccb74450d, sha256 0677efd401e73ee174bb2e30d650e8cf8b0f37fd7a3a2216245dbe0ecb0018fe)
BibTeX
@misc{dc043ccb74450d,
  title = {Automatic Lyric Transcription for Greek Songs: Scaling and Task Composition Effects in Whisper Adaptation — arXiv:2609.11302},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11302},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_043ccb74450d; sha256 0677efd401e73ee174bb2e30d650e8cf8b0f37fd7a3a2216245dbe0ecb0018fe},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11302. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Automatic Lyric Transcription (ALT) remains substantially more challenging than speech recognition due to melodic variability, rhythmic irregularity, and accompaniment interference. This is heightened in low-resource languages like Greek, where no prior benchmark for ALT exists. We present the first controlled study of Whisper adaptation for Greek ALT, investigating model scaling effects, task composition via multitask training in transcribe-translate ratios, and two-stage speech-to-singing adaptation. We also curate a segment-level aligned singing dataset based on the Greek Audio Dataset (GAD) using source separation and CTC forced alignment. Results show that scaling consistently improves performance, while multitask learning acts as a beneficial regularizer primarily for smaller-capacity models. The 2-stage adaptation in Whisper Large-v3 achieves a Word Error Rate (WER) of 27.2%, a significant improvement over zero-shot baselines, establishing the first Greek ALT benchmark.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Sound (cs.SD).
Pagina del lavoro: https://arxiv.org/abs/2609.11302 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T05:23:57.009Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).