TAU Urban Acoustic Scenes 2020 3Class: 10-Second Audio Segments from 10 Scenes

40 hours of 10-second audio segments from 10 acoustic scenes, grouped into three major classes: indoor, outdoor, and transportation. The development dataset was created by Toni Heittola of Tampere University for the TAU Urban Acoustic Scenes 2020 challenge.

Use Cases

• Train audio scene classifiers based on the three major scene classes (indoor, outdoor, transportation).
• Benchmark model performance for urban acoustic scene recognition tasks.
• Analyze acoustic characteristics of specific environments like airports, metro stations, and urban parks.

Strengths

• Contains 40 hours of total audio data.
• Audio segments are consistently 10 seconds in length.
• Scenes are logically grouped into three major classes for structured analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Tampere University

Collection Method

Likely consists of recorded or synthesized audio segments from urban environments.