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

arXiv:2609.11302v1 Announce Type: new Abstract: 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.

Related

• MoLGE: Mixture of Language Group Experts for Efficient Scaling of Massively Multilingual Speech Recognition
• Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts
• COALA: Robust Contextualized Speech-augmented Language Modeling for ASR via Contrastive Regularizer and Biasing Score Estimation

Source: arXiv cs.CL | 2026-09-11