An Analysis of Music Perception Skills on Crowdsourcing Platforms

An Analysis of Music Perception Skills on Crowdsourcing Platforms

Samiotis, I.P. (TU Delft Web Information Systems)
Qiu, S. (TU Delft Web Information Systems; Hunan Institute of Advanced Technology)
Lofi, C. (TU Delft Web Information Systems)
Yang, J. (TU Delft Web Information Systems)
Gadiraju, Ujwal (TU Delft Web Information Systems)
Bozzon, A. (TU Delft Web Information Systems; TU Delft Human-Centred Artificial Intelligence)

2022

Music content annotation campaigns are common on paid crowdsourcing platforms. Crowd workers are expected to annotate complex music artifacts, a task often demanding specialized skills and expertise, thus selecting the right participants is crucial for campaign success. However, there is a general lack of deeper understanding of the distribution of musical skills, and especially auditory perception skills, in the worker population. To address this knowledge gap, we conducted a user study (N = 200) on Prolific and Amazon Mechanical Turk. We asked crowd workers to indicate their musical sophistication through a questionnaire and assessed their music perception skills through an audio-based skill test. The goal of this work is to better understand the extent to which crowd workers possess higher perceptions skills, beyond their own musical education level and self reported abilities. Our study shows that untrained crowd workers can possess high perception skills on the music elements of melody, tuning, accent, and tempo; skills that can be useful in a plethora of annotation tasks in the music domain.

human computation
music annotation
perceptual skills
music sophistication
knowledge crowdsourcing

http://resolver.tudelft.nl/uuid:c27dd076-2e92-4d84-ae95-7e9720d8d5c8

https://doi.org/10.3389/frai.2022.828733

2624-8212

Frontiers in Artificial Intelligence, 5

Institutional Repository

journal article

© 2022 I.P. Samiotis, S. Qiu, C. Lofi, J. Yang, Ujwal Gadiraju, A. Bozzon