NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigms
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908276567834624 |
|---|---|
| author | Wang, Yashan Wu, Shangda Hu, Jianhuai Du, Xingjian Peng, Yueqi Huang, Yongxin Fan, Shuai Li, Xiaobing Yu, Feng Sun, Maosong |
| author_facet | Wang, Yashan Wu, Shangda Hu, Jianhuai Du, Xingjian Peng, Yueqi Huang, Yongxin Fan, Shuai Li, Xiaobing Yu, Feng Sun, Maosong |
| contents | We introduce NotaGen, a symbolic music generation model aiming to explore the potential of producing high-quality classical sheet music. Inspired by the success of Large Language Models (LLMs), NotaGen adopts pre-training, fine-tuning, and reinforcement learning paradigms (henceforth referred to as the LLM training paradigms). It is pre-trained on 1.6M pieces of music in ABC notation, and then fine-tuned on approximately 9K high-quality classical compositions conditioned on "period-composer-instrumentation" prompts. For reinforcement learning, we propose the CLaMP-DPO method, which further enhances generation quality and controllability without requiring human annotations or predefined rewards. Our experiments demonstrate the efficacy of CLaMP-DPO in symbolic music generation models with different architectures and encoding schemes. Furthermore, subjective A/B tests show that NotaGen outperforms baseline models against human compositions, greatly advancing musical aesthetics in symbolic music generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_18008 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigms Wang, Yashan Wu, Shangda Hu, Jianhuai Du, Xingjian Peng, Yueqi Huang, Yongxin Fan, Shuai Li, Xiaobing Yu, Feng Sun, Maosong Sound Artificial Intelligence Audio and Speech Processing We introduce NotaGen, a symbolic music generation model aiming to explore the potential of producing high-quality classical sheet music. Inspired by the success of Large Language Models (LLMs), NotaGen adopts pre-training, fine-tuning, and reinforcement learning paradigms (henceforth referred to as the LLM training paradigms). It is pre-trained on 1.6M pieces of music in ABC notation, and then fine-tuned on approximately 9K high-quality classical compositions conditioned on "period-composer-instrumentation" prompts. For reinforcement learning, we propose the CLaMP-DPO method, which further enhances generation quality and controllability without requiring human annotations or predefined rewards. Our experiments demonstrate the efficacy of CLaMP-DPO in symbolic music generation models with different architectures and encoding schemes. Furthermore, subjective A/B tests show that NotaGen outperforms baseline models against human compositions, greatly advancing musical aesthetics in symbolic music generation. |
| title | NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigms |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2502.18008 |