edm-cue

```
from datasets import load_dataset
captions = load_dataset("disco-eth/edm-cue")
```

# What is EDM-CUE?

The EDM-CUE dataset contains metadata for ~5k EDM tracks. Cue points are essential for DJs, so we asked the question "can they be placed by a learned system?" To Answer this question we gathered 21k cue points manually placed by human experts, and provide them in this dataset for future use.

To cite this dataset or for more information, please see [*Cue Point Estimation using Object Detection*](https://www.arxiv.org/abs/2407.06823).

## Dataset Structure

The dataset contains the following features:

* id: Track ID on Deezer.
* title: Track name.
* artists: Track artists.
* duration: Track duration in seconds.
* genre: List of genres (possibly incomplete).
* key: Alphanumeric key of the track generated by rekordbox (possibly incorrect).
* beat_grid: Contains 4 values to compute the grid. 'start_pos' is the starting position in seconds, 'init_beat' is the beat count of 'start_pos', 'bpm' is the beats per minute tempo, and 'time_sig' is the time signature in string format.
* cue_pts: Position of the cue point in seconds.

from datasets import load_dataset
captions = load_dataset("disco-eth/edm-cue")

# 什么是EDM-CUE？

EDM-CUE数据集收录了约5000首电子舞曲（EDM）曲目的元数据。节拍提示点（Cue Point）是DJ作业的核心依托要素，为此我们提出研究议题：能否借助学习型系统自动生成此类提示点？为解答该问题，我们收集了21000个由人类专家手动标注的节拍提示点，并将其纳入本数据集以供后续研究使用。

若需引用本数据集或获取更多相关信息，请参阅论文《基于目标检测的节拍提示点估计》（*Cue Point Estimation using Object Detection*），链接：https://www.arxiv.org/abs/2407.06823。

## 数据集结构

本数据集包含以下字段：
* `id`：Deezer平台上的曲目唯一标识符。
* `title`：曲目标题。
* `artists`：曲目参演艺人信息。
* `duration`：曲目时长，单位为秒。
* `genre`：曲目风格列表（可能存在不完整情况）。
* `key`：由rekordbox生成的曲目字母数字调号（可能存在误差）。
* `beat_grid`：用于计算节拍网格的四项参数。其中`start_pos`为网格起始位置（单位：秒），`init_beat`为该起始位置对应的节拍序号，`bpm`为每分钟节拍数（BPM），`time_sig`为字符串格式的拍号。
* `cue_pts`：节拍提示点的位置，单位为秒。