Dataset Description
This dataset includes 1,344 one-hour audio recordings from the homes of 8 participants in Belgium, captured during 2023. Each home was equipped with two AudioMoth devices, installed in the Living Room and Kitchen, recording simultaneously. The dataset is organized into 14 folders, each set of two folders representing audio recordings from the same house. The recordings document daily soundscapes and activities of living environments, offering a valuable resource for sound event detection research. For further details on the project's background, please visit our project page.
| Aspect | Details |
|---|---|
| Recording Device | AudioMoth with IPX7 Waterproof Case |
| Number of Samples | 1344 audio files |
| Length of Each File | 3595 seconds (approximately 1 hour) |
| Number of Homes | 7 |
| Recording Duration | 7 days per home |
| Recording Window | Typically 8:00 AM to 9:00 PM |
| Housing Types | Detached, semi-detached, mixed-use housing, high-rise flats |
| Location Types | Rural and suburban areas |
| Aspect | Details |
|---|---|
| Geography | Belgium (around Geel/Turnhout) |
| Demographics | 8 participants, all aged 55-80 years |
| Rooms Recorded | Primarily living rooms and kitchens |
| Floor Types | Varied: wood, linoleum, concrete |
| Wall Types | Varied: concrete, painted |
| Home Features Diversity | Large windows, soft furnishings, open plan designs, pets (dog and cat in one home), robot vacuum cleaner (in one home) |
| Special Notes | Two participants lived together in one house; Some homes had combined kitchen and living areas; Various noise issues noted (e.g., ventilation, ground-borne vibration) |
Recording Information
The following tables provide detailed information about the recorders installed in the kitchen and living areas, as well as their recording times.
Kitchen Recorders
| Recorder ID | Recording Times |
|---|---|
| Recorder 02 | 07:00, 08:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 04 | 06:30, 07:30, 08:30, 09:30, 10:30, 11:30, 12:30, 13:30, 14:30, 15:30, 16:30, 17:30, 18:30 |
| Recorder 05 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 07 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 10 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 11 | 07:00, 08:00, 09:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00, 23:00 |
| Recorder 13 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
Living Room Recorders
| Recorder ID | Recording Times |
|---|---|
| Recorder 01 | 07:00, 08:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 03 | 06:30, 07:30, 08:30, 09:30, 10:30, 11:30, 12:30, 13:30, 14:30, 15:30, 16:30, 17:30, 18:30 |
| Recorder 06 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 08 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 09 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
| Recorder 12 | 07:00, 08:00, 09:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00, 23:00 |
| Recorder 14 | 07:00, 08:00, 09:00, 10:00, 11:00, 12:00, 13:00, 14:00, 15:00, 16:00, 17:00, 18:00, 19:00, 20:00, 21:00, 22:00 |
Citation
If you use this dataset, please consider citing us:
Link to Arxiv Publication@article{bibbo2024sounds,
title={The Sounds of Home: A Speech-Removed Residential Audio Dataset for Sound Event Detection},
author={Bibbó, G. and Deacon, T. and Singh, A. and Plumbley, M. D.},
journal={CHiME 2024 workshop, Kos Island, Greece},
year={2024}
}
Download
The complete dataset is downloadable from the CVSSP server:
For segmented downloads, the dataset is also available in four parts on Zenodo:
The remaining parts can be found through links provided in Part 1.
Collaboration Institutions
University of Surrey and Licalab are the primary institutions involved in this project. Below are their logos:
Funding and Acknowledgments
This project has been supported by the Engineering and Physical Sciences Research Council (EPSRC) under the grant EP/T019751/1 "AI for Sound (AI4S)" and by the VITALISE Transnational Access programme.