Hugging Face's logo

Datasets:
soerenray
/
speech_commands_enriched_and_annotated

Modalities:
Audio
Text
Time-series
Formats:
parquet
Size:
10K - 100K
ArXiv:
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files
xet
Community
soerenray commited on
Commit
d644481
·
1 Parent(s): ed80063

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +13 -7
README.md CHANGED
@@ -2,8 +2,14 @@
2
  license: openrail
3
  dataset_info:
4
  features:
5
- - name: full_audio
6
- dtype: audio
 
 
 
 
 
 
7
  - name: label
8
  dtype: int64
9
  - name: is_unknown
@@ -23,13 +29,13 @@ dataset_info:
23
  - name: embedding
24
  sequence: float32
25
  - name: embedding_reduced
26
- sequence: float64
27
  splits:
28
  - name: train
29
- num_bytes: 1774663023.432
30
- num_examples: 51093
31
- download_size: 1701177850
32
- dataset_size: 1774663023.432
33
  ---
34
 
35
  This dataset is an extended version of the MIT/ast-finetuned-speech-commands-v2 dataset.
 
2
  license: openrail
3
  dataset_info:
4
  features:
5
+ - name: file
6
+ dtype: string
7
+ - name: audio
8
+ struct:
9
+ - name: bytes
10
+ dtype: binary
11
+ - name: path
12
+ dtype: 'null'
13
  - name: label
14
  dtype: int64
15
  - name: is_unknown
 
29
  - name: embedding
30
  sequence: float32
31
  - name: embedding_reduced
32
+ sequence: float32
33
  splits:
34
  - name: train
35
+ num_bytes: 2139150771
36
+ num_examples: 60973
37
+ download_size: 2095276737
38
+ dataset_size: 2139150771
39
  ---
40
 
41
  This dataset is an extended version of the MIT/ast-finetuned-speech-commands-v2 dataset.