Hugging Face's logo

Datasets:
DBD-research-group
/
BirdSet

Tasks:
Audio Classification
ArXiv:
Tags:
audio classification
multi-label classification
bird sounds
passive acoustic monitoring
DOI:
License:
Dataset card Files Community
2
lrauch commited on
Commit
0642cda
•
1 Parent(s): a289efd

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +25 -25
README.md CHANGED
@@ -47,20 +47,20 @@ We offer a static set of evaluation datasets and a varied collection of training
47
  - The bird species are translated to ebird_codes
48
  - Snapshot date of XC: 03/10/2024
49
 
50
- ##### Train
51
  - Exclusively using focal audio data from Xeno-Canto (XC) with quality ratings A, B, C and excluding all recordings that are CC-ND.
52
  - Each dataset is tailored for specific target species identified in the corresponding test soundscape files.
53
  - We transform the scientific names of the birds into the corresponding ebird_code label.
54
  - We offer detected events and corresponding cluster assignments to identify bird sounds in each recording.
55
  - We provide the full recordings from XC. These can generate multiple samples from a single instance.
56
 
57
- ##### Test_5s
58
  - Task: Multilabel ("ebird_code_multilabel")
59
  - Only soundscape data from Zenodo formatted acoording to the Kaggle evaluation scheme.
60
  - Each recording is segmented into 5-second intervals where each ground truth bird vocalization is assigned to.
61
  - This contains segments without any labels which results in a [0] vector.
62
 
63
- ##### Test
64
  - Task: Multiclass ("ebird_code")
65
  - Only soundscape data sourced from Zenodo.
66
  - We provide the full recording with the complete label set and specified bounding boxes.
@@ -75,28 +75,28 @@ We offer a static set of evaluation datasets and a varied collection of training
75
 
76
  | | format datasets. | description |
77
  |------------------------|-------------------------------------------------------:|-------------------------:|
78
- | audio | Audio(sampling_rate=32_000, mono=True, decode=True) | xxxxxx |
79
- | filepath | Value("string") | xxxxxx |
80
- | start_time | Value("float64") | xxxxxx |
81
- | end_time | Value("float64") | xxxxxx |
82
- | low_freq | Value("int64") | xxxxxx |
83
- | high_freq | Value("int64") | xxxxxx |
84
- | ebird_code | ClassLabel(names=class_list) | xxxxxx |
85
- | ebird_code_multilabel | Sequence(datasets.ClassLabel(names=class_list)) | x |
86
- | call_type | Sequence(datasets.Value("string")) | x |
87
- | sex | Value("string") | x |
88
- | lat | Value("float64") | x |
89
- | long | Value("float64") | x |
90
- | length | Value("int64") | x |
91
- | microphone | Value("string") | x |
92
- | license | Value("string") | x |
93
- | source | Value("string") | x |
94
- | local_time | Value("string") | x |
95
- | detected_events | Sequence(datasets.Sequence(datasets.Value("float64")))| x |
96
- | event_cluster | Sequence(datasets.Value("int64")) | x |
97
- | peaks | Sequence(datasets.Value("float64")) | x |
98
- | quality | Value("string") | x |
99
- | recordist | Value("string") | x |
100
 
101
  #### Example Metadata Train
102
 
 
47
  - The bird species are translated to ebird_codes
48
  - Snapshot date of XC: 03/10/2024
49
 
50
+ **Train**
51
  - Exclusively using focal audio data from Xeno-Canto (XC) with quality ratings A, B, C and excluding all recordings that are CC-ND.
52
  - Each dataset is tailored for specific target species identified in the corresponding test soundscape files.
53
  - We transform the scientific names of the birds into the corresponding ebird_code label.
54
  - We offer detected events and corresponding cluster assignments to identify bird sounds in each recording.
55
  - We provide the full recordings from XC. These can generate multiple samples from a single instance.
56
 
57
+ **Test_5s**
58
  - Task: Multilabel ("ebird_code_multilabel")
59
  - Only soundscape data from Zenodo formatted acoording to the Kaggle evaluation scheme.
60
  - Each recording is segmented into 5-second intervals where each ground truth bird vocalization is assigned to.
61
  - This contains segments without any labels which results in a [0] vector.
62
 
63
+ **Test**
64
  - Task: Multiclass ("ebird_code")
65
  - Only soundscape data sourced from Zenodo.
66
  - We provide the full recording with the complete label set and specified bounding boxes.
 
75
 
76
  | | format datasets. | description |
77
  |------------------------|-------------------------------------------------------:|-------------------------:|
78
+ | audio | Audio(sampling_rate=32_000, mono=True, decode=True) | |
79
+ | filepath | Value("string") | |
80
+ | start_time | Value("float64") | |
81
+ | end_time | Value("float64") | |
82
+ | low_freq | Value("int64") | |
83
+ | high_freq | Value("int64") | |
84
+ | ebird_code | ClassLabel(names=class_list) | |
85
+ | ebird_code_multilabel | Sequence(datasets.ClassLabel(names=class_list)) | |
86
+ | call_type | Sequence(datasets.Value("string")) | |
87
+ | sex | Value("string") | |
88
+ | lat | Value("float64") | |
89
+ | long | Value("float64") | |
90
+ | length | Value("int64") | |
91
+ | microphone | Value("string") | |
92
+ | license | Value("string") | |
93
+ | source | Value("string") | |
94
+ | local_time | Value("string") | |
95
+ | detected_events | Sequence(datasets.Sequence(datasets.Value("float64")))| |
96
+ | event_cluster | Sequence(datasets.Value("int64")) | |
97
+ | peaks | Sequence(datasets.Value("float64")) | |
98
+ | quality | Value("string") | |
99
+ | recordist | Value("string") | |
100
 
101
  #### Example Metadata Train
102