Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ pipeline_tag: audio-classification
|
|
| 11 |
---
|
| 12 |
|
| 13 |
# CultureMERT: Continual Pre-Training for Cross-Cultural Music Representation Learning
|
| 14 |
-
📑 [**Read the Paper**](TODO)
|
| 15 |
|
| 16 |
**CultureMERT-95M** is a multi-culturally adapted 95M-parameter music foundation model based on [MERT-v1-95M](https://huggingface.co/m-a-p/MERT-v1-95M). It is developed through a two-stage **continual pre-training** strategy on 650 hours of culturally diverse audio spanning Greek, Turkish, and Indian musical traditions. The model significantly improves representation quality for **"non-Western" music**, achieving an average ROC-AUC improvement of **+4.43%** across culturally diverse music tagging tasks, surpassing prior state-of-the-art, while maintaining strong performance on Western-centric benchmarks such as MagnaTagATune and FMA-medium.
|
| 17 |
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
# CultureMERT: Continual Pre-Training for Cross-Cultural Music Representation Learning
|
| 14 |
+
📑 [**Read the Paper**](...TODO)
|
| 15 |
|
| 16 |
**CultureMERT-95M** is a multi-culturally adapted 95M-parameter music foundation model based on [MERT-v1-95M](https://huggingface.co/m-a-p/MERT-v1-95M). It is developed through a two-stage **continual pre-training** strategy on 650 hours of culturally diverse audio spanning Greek, Turkish, and Indian musical traditions. The model significantly improves representation quality for **"non-Western" music**, achieving an average ROC-AUC improvement of **+4.43%** across culturally diverse music tagging tasks, surpassing prior state-of-the-art, while maintaining strong performance on Western-centric benchmarks such as MagnaTagATune and FMA-medium.
|
| 17 |
|