Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915670018490368 |
|---|---|
| author | Ou, Longshen Ma, Xichu Wang, Ye |
| author_facet | Ou, Longshen Ma, Xichu Wang, Ye |
| contents | Despite progress in melody-to-lyric generation, a substantial singability gap remains between machine-generated lyrics and those written by human lyricists. In this work, we aim to narrow this gap by jointly learning both wording and formatting for melody-to-lyric generation. After general-domain pretraining, our model acquires length awareness through an self-supervised stage trained on a large text-only lyric corpus. During supervised melody-to-lyric training, we introduce multiple auxiliary supervision objective informed by musicological findings on melody--lyric relationships, encouraging the model to capture fine-grained prosodic and structural patterns. Compared with naïve fine-tuning, our approach improves adherence to line-count and syllable-count requirements by 3.8% and 21.4% absolute, respectively, without degrading text quality. In human evaluation, it achieves 42.2% and 74.2% relative gains in overall quality over two task-specific baselines, underscoring the importance of formatting-aware training for generating singable lyrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_02146 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation Ou, Longshen Ma, Xichu Wang, Ye Computation and Language Sound Audio and Speech Processing Despite progress in melody-to-lyric generation, a substantial singability gap remains between machine-generated lyrics and those written by human lyricists. In this work, we aim to narrow this gap by jointly learning both wording and formatting for melody-to-lyric generation. After general-domain pretraining, our model acquires length awareness through an self-supervised stage trained on a large text-only lyric corpus. During supervised melody-to-lyric training, we introduce multiple auxiliary supervision objective informed by musicological findings on melody--lyric relationships, encouraging the model to capture fine-grained prosodic and structural patterns. Compared with naïve fine-tuning, our approach improves adherence to line-count and syllable-count requirements by 3.8% and 21.4% absolute, respectively, without degrading text quality. In human evaluation, it achieves 42.2% and 74.2% relative gains in overall quality over two task-specific baselines, underscoring the importance of formatting-aware training for generating singable lyrics. |
| title | Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2307.02146 |