agkphysics commited on
Commit
c051cda
1 Parent(s): d8c6eb3

Update README with additional metadata.

Browse files
Files changed (1) hide show
  1. README.md +30 -0
README.md CHANGED
@@ -1,3 +1,33 @@
1
  ---
2
  license: cc-by-4.0
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc-by-4.0
3
+ tags:
4
+ - audio
5
+ task_categories:
6
+ - audio-classification
7
  ---
8
+
9
+ # AudioSet data
10
+ This repository contains the balanced training set and evaluation set
11
+ of the [AudioSet data](
12
+ https://research.google.com/audioset/dataset/index.html). The YouTube
13
+ videos were downloaded in March 2023, and so not all of the original
14
+ audios are available.
15
+
16
+ Extracting the `*.tar` files will place audio clips into the `audio/`
17
+ directory. The distribuion of audio clips is as follows:
18
+ - `audio/bal_train`: 18685 audio clips out of 22160 originally.
19
+ - `audio/eval`: 17142 audio clips out of 20371 originally.
20
+
21
+ Most audio is sampled at 48 kHz 24 bit, but about 10% is sampled at
22
+ 44.1 kHz 24 bit. Audio files are stored in the FLAC format.
23
+
24
+ ## Citation
25
+ ```bibtex
26
+ @inproceedings{45857,
27
+ title = {Audio Set: An ontology and human-labeled dataset for audio events},
28
+ author = {Jort F. Gemmeke and Daniel P. W. Ellis and Dylan Freedman and Aren Jansen and Wade Lawrence and R. Channing Moore and Manoj Plakal and Marvin Ritter},
29
+ year = {2017},
30
+ booktitle = {Proc. IEEE ICASSP 2017},
31
+ address = {New Orleans, LA}
32
+ }
33
+ ```