D3RM: A Discrete Denoising Diffusion Refinement Model for Piano Transcription
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912185238683648 |
|---|---|
| author | Kim, Hounsu Kwon, Taegyun Nam, Juhan |
| author_facet | Kim, Hounsu Kwon, Taegyun Nam, Juhan |
| contents | Diffusion models have been widely used in the generative domain due to their convincing performance in modeling complex data distributions. Moreover, they have shown competitive results on discriminative tasks, such as image segmentation. While diffusion models have also been explored for automatic music transcription, their performance has yet to reach a competitive level. In this paper, we focus on discrete diffusion model's refinement capabilities and present a novel architecture for piano transcription. Our model utilizes Neighborhood Attention layers as the denoising module, gradually predicting the target high-resolution piano roll, conditioned on the finetuned features of a pretrained acoustic model. To further enhance refinement, we devise a novel strategy which applies distinct transition states during training and inference stage of discrete diffusion models. Experiments on the MAESTRO dataset show that our approach outperforms previous diffusion-based piano transcription models and the baseline model in terms of F1 score. Our code is available in https://github.com/hanshounsu/d3rm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_05068 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | D3RM: A Discrete Denoising Diffusion Refinement Model for Piano Transcription Kim, Hounsu Kwon, Taegyun Nam, Juhan Sound Artificial Intelligence Machine Learning Audio and Speech Processing Diffusion models have been widely used in the generative domain due to their convincing performance in modeling complex data distributions. Moreover, they have shown competitive results on discriminative tasks, such as image segmentation. While diffusion models have also been explored for automatic music transcription, their performance has yet to reach a competitive level. In this paper, we focus on discrete diffusion model's refinement capabilities and present a novel architecture for piano transcription. Our model utilizes Neighborhood Attention layers as the denoising module, gradually predicting the target high-resolution piano roll, conditioned on the finetuned features of a pretrained acoustic model. To further enhance refinement, we devise a novel strategy which applies distinct transition states during training and inference stage of discrete diffusion models. Experiments on the MAESTRO dataset show that our approach outperforms previous diffusion-based piano transcription models and the baseline model in terms of F1 score. Our code is available in https://github.com/hanshounsu/d3rm. |
| title | D3RM: A Discrete Denoising Diffusion Refinement Model for Piano Transcription |
| topic | Sound Artificial Intelligence Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2501.05068 |