Update README.md
Browse files
README.md
CHANGED
@@ -6,7 +6,7 @@ tags:
|
|
6 |
- music
|
7 |
---
|
8 |
# MelodyT5: A Unified Score-to-Score Transformer for Symbolic Music Processing [ISMIR 2024]
|
9 |
-
This repository contains the code for the MelodyT5 model as described in the paper [MelodyT5: A Unified Score-to-Score Transformer for Symbolic Music Processing](https://arxiv.org/abs/
|
10 |
|
11 |
MelodyT5 is an unified framework for symbolic music processing, using an encoder-decoder architecture to handle multiple melody-centric tasks, such as generation, harmonization, and segmentation, by treating them as score-to-score transformations. Pre-trained on [MelodyHub](https://huggingface.co/datasets/sander-wood/melodyhub), a large dataset of melodies in ABC notation, it demonstrates the effectiveness of multi-task transfer learning in symbolic music processing.
|
12 |
|
@@ -492,7 +492,7 @@ K:D
|
|
492 |
fed dB/c/d | efe efg | fed daa | agf eag | fed B2 d | A2 d F2 A | Bdd F2 E | FDD D2 :|
|
493 |
```
|
494 |
|
495 |
-
## BibTeX
|
496 |
```
|
497 |
@misc{wu2024language,
|
498 |
title={Beyond Language Models: Byte Models are Digital World Simulators},
|
@@ -502,4 +502,4 @@ fed dB/c/d | efe efg | fed daa | agf eag | fed B2 d | A2 d F2 A | Bdd F2 E | FDD
|
|
502 |
archivePrefix={arXiv},
|
503 |
primaryClass={cs.LG}
|
504 |
}
|
505 |
-
```
|
|
|
6 |
- music
|
7 |
---
|
8 |
# MelodyT5: A Unified Score-to-Score Transformer for Symbolic Music Processing [ISMIR 2024]
|
9 |
+
This repository contains the code for the MelodyT5 model as described in the paper [MelodyT5: A Unified Score-to-Score Transformer for Symbolic Music Processing](https://arxiv.org/abs/2407.02277).
|
10 |
|
11 |
MelodyT5 is an unified framework for symbolic music processing, using an encoder-decoder architecture to handle multiple melody-centric tasks, such as generation, harmonization, and segmentation, by treating them as score-to-score transformations. Pre-trained on [MelodyHub](https://huggingface.co/datasets/sander-wood/melodyhub), a large dataset of melodies in ABC notation, it demonstrates the effectiveness of multi-task transfer learning in symbolic music processing.
|
12 |
|
|
|
492 |
fed dB/c/d | efe efg | fed daa | agf eag | fed B2 d | A2 d F2 A | Bdd F2 E | FDD D2 :|
|
493 |
```
|
494 |
|
495 |
+
<!-- ## BibTeX
|
496 |
```
|
497 |
@misc{wu2024language,
|
498 |
title={Beyond Language Models: Byte Models are Digital World Simulators},
|
|
|
502 |
archivePrefix={arXiv},
|
503 |
primaryClass={cs.LG}
|
504 |
}
|
505 |
+
``` -->
|