teticio/audio-diffusion-512
收藏Hugging Face2023-06-19 更新2024-03-04 收录
下载链接：
https://hf-mirror.com/datasets/teticio/audio-diffusion-512
下载链接
链接失效反馈官方服务：
资源简介：
---
dataset_info:
features:
- name: image
dtype: image
- name: audio_file
dtype: string
- name: slice
dtype: int16
splits:
- name: train
num_bytes: 1903861364.293
num_examples: 10663
download_size: 1903696036
dataset_size: 1903861364.293
pretty_name: Mel spectrograms of music
size_categories:
- 10K<n<100K
source_datasets: []
tags:
- audio
- spectrograms
task_categories:
- image-to-image
task_ids: []
---
Over 20,000 512x512 mel spectrograms of 5 second samples of music from my Spotify liked playlist. The code to convert from audio to spectrogram and vice versa can be found in https://github.com/teticio/audio-diffusion along with scripts to train and run inference using De-noising Diffusion Probabilistic Models.
```
x_res = 512
y_res = 512
sample_rate = 22050
n_fft = 2048
hop_length = 512
```
提供机构：
teticio
原始信息汇总
数据集概述
数据集基本信息
- 名称: Mel spectrograms of music
- 大小: 10K<n<100K
- 标签:
- audio
- spectrograms
- 任务类别: image-to-image
数据集特征
- image: 图像数据类型
- audio_file: 字符串数据类型
- slice: 整数数据类型（int16）
数据集分割
- 训练集:
- 示例数量: 10663
- 数据大小: 1903861364.293字节
数据集下载与大小
- 下载大小: 1903696036字节
- 数据集总大小: 1903861364.293字节
数据集详细参数
- x_res: 512
- y_res: 512
- sample_rate: 22050
- n_fft: 2048
- hop_length: 512
