Diversity and bias in audio captioning datasets

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

5 Downloads (Pure)


Describing soundscapes in sentences allows better understanding of the acoustic scene than a single label indicating the acoustic scene class or a set of audio tags indicating the sound events active in the audio clip. In addition, the richness of natural language allows a range of possible descriptions for the same acoustic scene. In this work, we address the diversity obtained when collecting descriptions of soundscapes using crowdsourcing. We study how much the collection of audio captions can be guided by the instructions given in the annotation task, by analysing the possible bias introduced by auxiliary information provided in the annotation process. Our study shows that even when given hints on the audio content, different annotators describe the same soundscape using different vocabulary. In automatic captioning, hints provided as audio tags represent grounding textual information that facilitates guiding the captioning output towards specific concepts. We also release a new dataset of audio captions and audio tags produced by multiple annotators for a subset of the TAU Urban Acoustic Scenes 2018 dataset, suitable for studying guided captioning.
Original languageEnglish
Title of host publicationProceedings of the 6th Workshop on Detection and Classication of Acoustic Scenes and Events (DCASE 2021)
EditorsFrederic Font, Annamaria Mesaros, Daniel P.W. Ellis, Eduardo Fonseca, Magdalena Fuentes, Benjamin Elizalde
ISBN (Electronic)978-84-09-36072-7
Publication statusPublished - 15 Nov 2021
Publication typeA4 Article in conference proceedings
EventDetection and Classication of Acoustic Scenes and Events - , Spain
Duration: 15 Nov 202119 Nov 2021


ConferenceDetection and Classication of Acoustic Scenes and Events

Publication forum classification

  • Publication forum level 0


Dive into the research topics of 'Diversity and bias in audio captioning datasets'. Together they form a unique fingerprint.

Cite this