Using State of the Art Technology (including Speech Analysis) to Reduce the Cost of Captioning/Subtitling Creation and Repurposing

摘要

Broadcasters today must support morechannels and formats on a wide variety of platformssuch as the Web, iPads/tablets and mobile phones,and within very tight budgets. Meanwhile, importantmandates for content accessibility (such as theAccessibility Act in the United States) are gettingmore stringent and include emerging platforms.Furthermore, to optimize revenue, broadcasters aredelivering more multi-language channels usingcaptioning and subtitling to reach a wider audience.Cost-effective captioning and subtitling is a keychallenge for broadcasters evolving to meet thesedemands. One of the most time-consuming elementsin the captioning/subtitling process is thesynchronization of captions and subtitles to the videocontent. Speech analysis technology is nowintegrated into the latest captioning and subtitlingtools to free operators from manually setting thetiming for each caption and subtitle. Using thedialogue from the audio track, shot changeinformation, reading speed calculations and otherconsiderations, the new solutions instantly assigntimecodes to captions and subtitles. Early tests showthat these tools can reduce creation time by as muchas 40%, and re-purposing by up to 90%.Once created, the captions must be presented tothe viewer. This requires a modified workflow toensure the various platform capabilities anddeficiencies are fully accounted for such that thecaptions arrive on time and in the correct format.