admin
commited on
Commit
·
bd04398
1
Parent(s):
7b5a3f8
upd md
Browse files
README.md
CHANGED
|
@@ -15,21 +15,20 @@ tags:
|
|
| 15 |
Our evaluation methodology adopted the approach for structural segmentation evaluation outlined in the Harmonix set, which employed Structural Features for boundary identification, and 2D-Fourier Magnitude Coefficients (2D-FMC) for segment labeling based on acoustic similarity. CQT features serve as input features for the algorithm. The algorithm is implemented using Music Structure Analysis Framework (MSAF). For evaluation metrics, the F-measure is reported for the following metrics: Hit Rate with 0.5 and 3-second windows for boundary retrieval, Pairwise Frame Clustering and Entropy Scores for segment labeling. The evaluation is implemented using mir_eval.
|
| 16 |
|
| 17 |
## Evaluation result
|
| 18 |
-
|
| 19 |
|
| 20 |
-
##
|
| 21 |
-
### By Git
|
| 22 |
-
```bash
|
| 23 |
-
git clone https://www.modelscope.cn/ccmusic-database/song_structure.git
|
| 24 |
-
pip install modelscope
|
| 25 |
-
```
|
| 26 |
-
|
| 27 |
-
### By API
|
| 28 |
```python
|
| 29 |
from modelscope import snapshot_download
|
| 30 |
model_dir = snapshot_download('ccmusic-database/song_structure')
|
| 31 |
```
|
| 32 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
## Dataset
|
| 34 |
<https://huggingface.co/datasets/ccmusic-database/song_structure>
|
| 35 |
|
|
@@ -37,17 +36,4 @@ model_dir = snapshot_download('ccmusic-database/song_structure')
|
|
| 37 |
<https://www.modelscope.cn/models/ccmusic-database/song_structure>
|
| 38 |
|
| 39 |
## Evaluation
|
| 40 |
-
<https://github.com/monetjoe/ccmusic_eval/tree/msa>
|
| 41 |
-
|
| 42 |
-
## Cite
|
| 43 |
-
```bibtex
|
| 44 |
-
@dataset{zhaorui_liu_2021_5676893,
|
| 45 |
-
author = {Monan Zhou, Shenyang Xu, Zhaorui Liu, Zhaowen Wang, Feng Yu, Wei Li and Baoqiang Han},
|
| 46 |
-
title = {CCMusic: an Open and Diverse Database for Chinese and General Music Information Retrieval Research},
|
| 47 |
-
month = {mar},
|
| 48 |
-
year = {2024},
|
| 49 |
-
publisher = {HuggingFace},
|
| 50 |
-
version = {1.2},
|
| 51 |
-
url = {https://huggingface.co/ccmusic-database}
|
| 52 |
-
}
|
| 53 |
-
```
|
|
|
|
| 15 |
Our evaluation methodology adopted the approach for structural segmentation evaluation outlined in the Harmonix set, which employed Structural Features for boundary identification, and 2D-Fourier Magnitude Coefficients (2D-FMC) for segment labeling based on acoustic similarity. CQT features serve as input features for the algorithm. The algorithm is implemented using Music Structure Analysis Framework (MSAF). For evaluation metrics, the F-measure is reported for the following metrics: Hit Rate with 0.5 and 3-second windows for boundary retrieval, Pairwise Frame Clustering and Entropy Scores for segment labeling. The evaluation is implemented using mir_eval.
|
| 16 |
|
| 17 |
## Evaluation result
|
| 18 |
+

|
| 19 |
|
| 20 |
+
## Usage
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
```python
|
| 22 |
from modelscope import snapshot_download
|
| 23 |
model_dir = snapshot_download('ccmusic-database/song_structure')
|
| 24 |
```
|
| 25 |
|
| 26 |
+
## Maintenance
|
| 27 |
+
```bash
|
| 28 |
+
git clone git@hf.co:ccmusic-database/song_structure
|
| 29 |
+
cd song_structure
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
## Dataset
|
| 33 |
<https://huggingface.co/datasets/ccmusic-database/song_structure>
|
| 34 |
|
|
|
|
| 36 |
<https://www.modelscope.cn/models/ccmusic-database/song_structure>
|
| 37 |
|
| 38 |
## Evaluation
|
| 39 |
+
<https://github.com/monetjoe/ccmusic_eval/tree/msa>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|