Lost in Transcription? Captioning Music and Sound in Peak Television

On October 9, 2012, Netflix entered into a landmark consent decree with the National Association of the Deaf that, by October, 2014, captions “will be available on 100% of [its] on-demand streaming content”. Hailed by access advocates, the practice enabled D/deaf and hard-of-hearing Americans to enjoy Netflix programming, which necessarily transcribes “all spoken dialogue”. For other soundtrack elements, however, the agreement only required captions to “contain descriptors, identifying important non-dialogue sounds”. This essentially gave license to the company’s army of freelance captioners to decide which sounds—including music—are “important” and how to describe them. Thus the sound effects and music cues in Netflix content are subject to the knowledge of these uncredited gig-economy captioners, whose “descriptors” run the gamut of detail and accuracy, not least because of the exploitive conditions under which most of them work. This chapter intends to open the study of music captions in streaming television by focusing on practices on one media platform, Netflix. Since 2014, Netflix has operated under its own “vococentric” guidelines for captioning—even though they represent the most detailed of any streaming service, their policies nevertheless leave considerable latitude for describing sound events, especially music. The study begins with a discussion of terminology, history, and practices for captioning music in audiovisual media. It is followed by a granular study of closed captioning of music for recent seasons of the series Vikings: Valhalla, Archive 81, and Stranger Things, which reveals the variability and differences in practices. Supported by published accounts of and interviews with workers in the industry, the chapter exposes a little-known industry and its practices related to the captioning of music, which has become so important for the Peak television experience of a growing non-Deaf and D/deaf audience.