File size: 2,742 Bytes
3e94add
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
---
license: cc-by-4.0
task_categories:
- audio-classification
language:
- en
pretty_name: DataSEC v1
tags:
- audio
- sound-event-classification
- environmental-sound
- urban-sound
size_categories:
- 1K<n<10K
configs:
- config_name: default
  data_files:
  - split: train
    path:
    - metadata.csv
    - PT_DATASET_250314/**/*.wav
  drop_labels: false
---

# DataSEC v1

This is a Hugging Face mirror of the CC-BY-4.0 v1 DataSEC archive from Zenodo record [10.5281/zenodo.15340689](https://doi.org/10.5281/zenodo.15340689).

DataSEC is a sound event classification dataset for environmental noise. The files are organized by class, with selected classes further organized by subclass.

Important version note: the latest Zenodo version of DataSEC uses `cc-by-nc-sa-4.0`, so this repository intentionally mirrors the earlier `PT_DATASET_250314.zip` v1 archive, whose Zenodo metadata is `cc-by-4.0`.

## Verified Archive Statistics

The v1 Zenodo description text mentions 5,024 samples and 23 hours 32 minutes. The actual `PT_DATASET_250314.zip` archive downloaded from the CC-BY-4.0 v1 record contains the following verified contents:

| Item | Value |
| --- | ---: |
| WAV files | 3,226 |
| Top-level classes | 22 |
| Subclasses | 15 |
| Total duration | 15.27 hours |
| Audio format | WAV, PCM 16-bit |
| Sample rate | 44,100 Hz |
| Channels | Mono |

## Files

- `PT_DATASET_250314/`: audio files organized by class and subclass.
- `metadata.csv`: metadata for all WAV files, including label, subclass, duration, sample rate, channels, bit depth, and frame count.
- `stats.json`: machine-readable statistics generated from the extracted archive.

## Class Counts

| Class | Files |
| --- | ---: |
| Bells | 67 |
| Birds | 69 |
| Cat fights and moans | 50 |
| Chicken coop | 58 |
| Cicadas and crickets | 66 |
| Crows seagulls and magpies | 101 |
| Dog barkings and howlings | 70 |
| Glass breaking | 109 |
| Horn | 57 |
| Jet aircrafts | 103 |
| Lawn mower brush cutter and olive shaker | 116 |
| Music | 102 |
| Propeller aircrafts | 108 |
| Sirens and alarms | 105 |
| Thunder fireworks and gunshot | 235 |
| Train | 67 |
| Vacuum cleaner fan and hairdryer | 95 |
| Vehicle idling | 112 |
| Vehicle pass-by | 217 |
| Voices | 1,001 |
| Wind turbine | 100 |
| Workshop | 218 |

## Source And Checksum

Source record: [Zenodo 15340689](https://zenodo.org/records/15340689)

Verified upstream MD5 checksum:

- `PT_DATASET_250314.zip`: `1fdc062919d70f43e1864e781b6a2232`

## Citation

```bibtex
@dataset{datasec_v1_2025,
  title = {DataSEC - Dataset for Sound Event Classification of environmental noise},
  year = {2025},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.15340689},
  url = {https://doi.org/10.5281/zenodo.15340689}
}
```