WASHINGTON—A recent series of FCC amendments and clarifications regarding online closed captioning of broadcast content runs nearly half the length of the original 102-page FCC regulation. The rules implement provisions of the “Twenty-First Century Communications & Video Accessibility Act of 2010,” as broadcasters, device makers, software providers and consumers await a series of additional deadlines and, no doubt, further adjustments by the commission.
The FCC released its “Order on Reconsideration and Further Notice of Proposed Rulemaking” in June, which modified and clarified its original Report and Order. The rules, which took effect in September 2012, primarily covered full-length TV programming, but starting the end of next month (Sept. 30), all pre-recorded video programming that is “substantially edited” for the Internet must be captioned if it was shown on television with captions, according to the FCC. (Changing the number or duration of commercial spots is not considered “substantially editing.”)
At the same time, the manufacturing deadline for video devices to be able to process closed captioning is not until Jan. 1, 2014 (and will be extended for Blu-ray and DVD players). But such devices will not be required to actually be available to consumers by the deadline. DSLRs are exempt, and most TV video “clips” and outtakes streamed online also will not require captioning.
“Unlike broadcast video, where every station transmits the same ATSC spec and every consumer TV set can display the closed captions carried in ATSC video, the Web is like the ‘Wild Wild West’ of video formats,” said Jason Livingston, project manager and developer at CPC, a closed- caption software firm in Rockville, Md. In late July, Livingston conducted a Webcast updating closed-captioning regs under the auspices of the Society of Motion Picture & Television Engineers. The organization’s SMPTE Timed Text [PDF] (SMPTE 2052) format strives to unite the myriad formats for captioning, according to Livingston.
“The idea is that regardless of what Web video delivery technology—HTML5, Flash, Silverlight, HLS, HDS, MPEG DASH, and so on—and what playback device you’re using, SMPTE Timed Text can be the captioning format for all these formats,” Livingston said. “However, at this time most captioners are not authoring directly in SMPTE-TT, and most browsers, mobile devices, and such are not using SMPTE-TT as a playback format. Instead, many have standardized on using SMPTE-TT as a ‘mezzanine’ format—a sort of ‘in-between’ used as a template for converting to all the different formats.”
Speaking solely for CPC, Livingston said there have been a lot of online content providers and distribution companies who have “expressed concerns” about the new regulations. “One of the things that makes captioning workflows difficult is that caption data is fragile and can be lost at any step in the process—content creation, editing, transcoding, distribution, hosting, and online delivery,” he added. “For captioning to work end-to-end, not only do all of these tools have to support closed captioning, but they all have to be mutually compatible in terms of how they work with captioning.”
Livingston said since content owners already have large libraries of captioned content in various broadcast formats, extracting existing captions allows broadcasters to re-use such data for their VOD online content.
“CPC added a new feature to our software called ‘QCC Multi-format Import,’” he said. “This can extract closed-captioning data from the majority of file-types used by broadcasters, using only software. The captions can then be automatically converted into formats like SMPTE-TT without human interaction. We also introduced a technology called ‘Assemble Captions’ which can automatically conform original captions to match the EDL [edit decision list] from the nonlinear editing software.”
Bill McLaughlin, software systems manager at EEG Enterprises in Farmingdale, N.Y., also said repurposing existing broadcast data as simply as feasible is a priority—ideally while working with a broadcaster’s existing distribution framework. EEG has several caption-authoring and extraction tools—notably CCRecord, CaptionTrack, and CCPlay FilePro for this purpose.
“The CC file extraction can be done during a live broadcast or offline in file-based or linear,” said McLaughlin. “EEG can also directly uplink real-time closed-caption data chunks from our HD490 SDI encoders or DE285 decoders to remote video servers over HTTP or Flash.”
According to Ken Frommert, marketing director for ENCO in Southfield, Mich., the original thought when the company launched its enCaption series several years ago was that any future FCC mandate would create a market for products that allow stations to remain compliant without investing a significant amount of money into captioning.
“ENCO recognizes the importance of captioning a live broadcast—especially in places where prompter feeds fall short of truly serving the hearing-impaired audience,” Frommert said. “Technologies like our enCaption3 are now reliable enough to be a viable [lower cost] alternative for most broadcasters. Even in places where enCaption3 doesn’t make sense for 24/7 use, it still fills a significant hole for those looking to caption breaking news or have a back-up solution ready to go,” Frommert said. ENCO’s enCaption3 automated system uses an enhanced speech-recognition engine to deliver closed captions in near real-time. (enCaption3 is also available as a monthly lease in nearly two dozen languages.)
Darren Forster, director of product management, subtitles and captioning initiatives for Miranda (which acquired U.K.-based captioning provider Softel earlier this year), believes that the FCC’s inclusion of the “safe harbor” format of SMPTE-TT for interchange and delivery in its closed caption mandate was crucial.
“Our experience, thus far, is that most distribution chains are using a dual-format approach, where closed-caption data is sent as both a separate SMPTE-TT file and embedded within the media,” he said. “This satisfies the [FCC] mandate and means that vendor interoperability issues are less likely.”
Miranda/Softel’s Swift Create V8 in its Swift Create Caption and Subtitling series, provides more than 20 new functions to simplify the captioning process.
“The main challenge of multiplatform captioning is there can be small variations in content, format and video duration for each use case, and thus, the caption files need to be re-purposed to fit,” Forster added. “Swift Create V8 offers tools to automate caption deletion after editing and to provide side-by-side caption comparisons [for] reduced re-purposing costs.” Softel’s Swift QC also is a new software tool designed to speed up re-purposing of caption-laden broadcast video.