Multi-Scene Noise Recordings From Daily Life

Featuring audio recordings of noise from various daily life scenes, including subway, supermarket, restaurant, road, and airport. The data was recorded using a Sony ICD-UX560F professional recorder in a high sampling rate, two-channel format. The author is Nexdata, and the dataset was last updated in April 2025.

Use Cases

• Train an audio event detection model to classify noise from specific scenes like subway or airport.
• Analyze the spectral characteristics of two-channel audio recordings from different environments like restaurants or highways.
• Use the high-quality recordings as background noise for augmenting speech datasets in machine learning pipelines.
• Compare the acoustic profiles of controlled indoor spaces like cinemas against outdoor scenes like city roads.

Strengths

• Recordings were made with a professional Sony ICD-UX560F recorder, indicating high-fidelity capture.
• Audio is recorded in a high sampling rate and two-channel format, providing detailed sound data.
• Covers a variety of daily life scenes including subway, supermarket, restaurant, road, and airport.

Limitations

• The dataset is described as a sample of a larger paid dataset, so the available scope is limited.
• Key metadata such as total number of recordings, file formats, and exact sampling rate is unknown.
• Geographic origin and specific time range of the recordings are not provided, limiting contextual analysis.

Provenance

Source

Nexdata

Collection Method

Recorded using a Sony ICD-UX560F professional recorder in a high sampling rate, two-channel format.

Freshness

Last updated on 2025-04-23.