Datasets:

Size Categories:
100K<n<100M
Language Creators:
unknown
Annotations Creators:
unknown
Source Datasets:
unknown
License:
Fhrozen commited on
Commit
4c99779
1 Parent(s): c739ab6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +31 -0
README.md CHANGED
@@ -51,6 +51,37 @@ The AudioSet ontology is a collection of sound events organized in a hierarchy.
51
 
52
  **This repository only includes audio files for DCASE 2022 - Task 3**
53
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
54
  ### Supported Tasks and Leaderboards
55
 
56
  - `audio-classification`: The dataset can be used to train a model for Sound Event Detection/Localization.
 
51
 
52
  **This repository only includes audio files for DCASE 2022 - Task 3**
53
 
54
+ The included labels are limited to:
55
+ - Female speech, woman speaking
56
+ - Male speech, man speaking
57
+ - Clapping
58
+ - Telephone
59
+ - Telephone bell ringing
60
+ - Ringtone
61
+ - Laughter
62
+ - Domestic sounds, home sounds
63
+ - Vacuum cleaner
64
+ - Kettle whistle
65
+ - Mechanical fan
66
+ - Walk, footsteps
67
+ - Door
68
+ - Cupboard open or close
69
+ - Music
70
+ - Background music
71
+ - Pop music
72
+ - Musical instrument
73
+ - Acoustic guitar
74
+ - Marimba, xylophone
75
+ - Cowbell
76
+ - Piano
77
+ - Electric piano
78
+ - Rattle (instrument)
79
+ - Water tap, faucet
80
+ - Bell
81
+ - Bicycle bell
82
+ - Chime
83
+ - Knock
84
+
85
  ### Supported Tasks and Leaderboards
86
 
87
  - `audio-classification`: The dataset can be used to train a model for Sound Event Detection/Localization.