NIPS4Bplus: a richly annotated birdsong audio dataset

View article
PeerJ Computer Science

Main article text

 

Introduction

Audio Data Collection

Annotations

Tags

Temporal annotations

  • The original tags were used for guidance; however, some files were judged to have a different set of species than the ones given in the original metadata. Similarly, in a few rare occurrences, despite the tags suggesting a bird species active in a recording, the annotator was not able to detect any bird vocalisation.

  • An extra ‘Unknown’ tag was added to the dataset for vocalisations that could not be classified to a class.

  • An extra ‘Human’ tag was added to a few recordings that have very obvious human sounds, such as speech, present in them.

  • Out of the 687 recordings of the training set 100 recordings contain only background noise, hence no temporal annotations were needed for them.

  • Of the remaining 587 recordings that contain vocalisations, six could not be unambiguously labelled due to hard to identify vocalisations, thus no temporal annotation files were produced for them.

  • An annotation file for any recording containing multiple insects does not differentiate between the insect species and the ‘Unknown’ label was given to all insect species present.

  • In the rare case where no birds were active along with the insects no annotation file was provided. Hence, seven recordings containing only insects were left unlabelled.

  • In total, 13 recordings have no temporal annotation files. These can be used when training a model that does not use temporal annotations.

  • On some occasions, the different syllables of a song were separated in time into different events while in other occasions they were summarised into a larger event, according to the judgement of the expert annotator. This variety could help train an unbiased model regarding separating events or grouping them together as one continuous time event.

Example Uses of NIPS4Bplus

Conclusion

Additional Information and Declarations

Competing Interests

Dan Stowell is an Academic Editor for PeerJ.

Author Contributions

Veronica Morfi conceived and designed the experiments, performed the experiments, analyzed the data, contributed reagents/materials/analysis tools, prepared figures and/or tables, performed the computation work, authored or reviewed drafts of the paper, approved the final draft.

Yves Bas, Hanna Pamuła and Hervé Glotin analyzed the data, contributed reagents/materials/analysis tools, authored or reviewed drafts of the paper, approved the final draft.

Dan Stowell conceived and designed the experiments, analyzed the data, contributed reagents/materials/analysis tools, prepared figures and/or tables, authored or reviewed drafts of the paper, approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The data is available at Figshare: Morfi, Veronica; Stowell, Dan; Pamula, Hanna (2018): NIPS4Bplus: Transcriptions of NIPS4B 2013 Bird Challenge Training Dataset. figshare. Dataset. https://doi.org/10.6084/m9.figshare.6798548.

Funding

Dan Stowell is supported by EPSRC fellowship EP/L020505/1. Hanna Pamuła is supported by AGH-UST Dean’s Grant number 16.16.130.942. SABIOD MI CNRS provided financial support for the NIPS4B challenge, and EADM MaDICS CNRS provided ANR-18-CE40-0014 SMILES supporting this research.

16 Citations 3,650 Views 2,007 Downloads

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more