File size: 569 Bytes
bce7b51
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
---
annotations_creators: []
language: []
language_creators: []
license: []
multilinguality: []
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 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 Denoising Diffusion Probabilistic Models.