Zebra Finch Pair Recordings with Annotated Vocal Segments

One 7-minute audio recording file per mixed-sex zebra finch pair, containing animal-borne accelerometer, microphone, and video channels. The dataset includes manually annotated vocal segments and was published as an appendix to a 2022 paper by researchers at ETH Zurich.

Use Cases

• Train vocalization detection models based on manually annotated audio segments.
• Analyze correlations between vocalizations and movement using synchronized accelerometer data.
• Study multimodal animal behavior patterns based on combined audio, video, and accelerometer streams.

Strengths

• Manually annotated vocal segments provide a high-quality ground truth.
• Multimodal data includes synchronized audio, video, and accelerometer channels from animal-borne sensors.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

ETH Zurich

Collection Method

Recordings of freely behaving zebra finch pairs in the BirdPark.