Abstract

ABSTRACT:

During the height of the COVID-19 pandemic, classical musicians had to find virtual alternatives for reaching their audiences after canceling recitals and concert seasons. Social media platforms seemingly provided the perfect solution by providing the ability to broadcast live music and upload prerecorded performances for audiences to view, leading to income and a sustained social media presence; however, the use of hash matching and classification machine learning to monitor uploaded content for copyright infringement complicates the issue when these methods are unable to distinguish between copyrighted performances and original recordings of works in the public domain. This article explains the mechanics and applications of hashing and classification methods in YouTube and Facebook's automated copyright enforcement systems and suggests incorporating crowdsourcing, voice recognition technology, metadata analysis, and human moderator intervention as improvements which may mitigate the negative impacts these programs may have on classical musicians.

pdf

Share