Submitted by Kate Pourshariati, Montgomery County Community College

Submitted by Kate Pourshariati, Montgomery County Community College

Closed Captioning session,“Capturing Captioning:

Problems in Preservation and Presentation of Timed Text” AMIA Pittsburgh 2016

The session opened with an overview of the history of captioning, while mentioning that there are circa 38 million deaf and hearing impaired media viewers in the United States. The captions also provide access to ESL students, and others with difficulties following spoken text.

The National Technical Institute for the Deaf or NTID is located at RIT in Rochester where they have a large integrated deaf community, they have lead the way on several technical innovations beginning in 1968. It may be interesting to check their deaf studies archives.

There was an in-depth discussion of the attempt to capture captions from older videotapes that originally had captions embedded. That information is captured in the video signal line 21. Sometimes this information can be lost when the tape is converted to a .mov file. What needs to happen apparently is the video needs to initially be captured uncompressed, and that file can then be converted to a smaller (.movetc) file.

The panelist whose work most mirrored what most of us do was Carleton Jackson of the University of Maryland. I have provided data from some of his power point slides below.* In general he said that they rip DVDs and otherwise convert files using Adobe Media Encoder and a modified version of Handbrake (open source). They have used a freeware package called Jubler (after using Notepad to create the primary text) to create new subtitles where they do not exist already, creating an SRT file for that purpose.

One interesting part was that U.M. accessesSubscene—which is a crowd-sourced captioning site, there are titles many already captioned feature films there, created by community members. Users are uploading SRT files to this site from all around the world. As an example, I was looking at Singing in the Rain,theArabic subtitles, there are many languages there besides English. [Perhaps it would be a good idea to share SRT files on Subscene so that we are not duplicating each others’ captioning work.] See also: 10 sites for you to download movie subtitles

One of the most important things that we learned about is the work being done by NCAM at WGBH Boston.The NCAM website reveals a newly funded initiative to create software to make it easy for everyone to create subtitles, possibly bringing in voice recognition. [These efforts are made complicated by films which have more than one language spoken, or have accents that are distinctive]. There is earlier application called MAGpie at the site for subtitling which perhaps can act as aQuiktime plug in.

[It was not discussed at this session, but one of the simplest platforms for closed captioning is Youtube itself. One would need to have the ripped file and either a prepared SRT file or create the titles live. Other Youtube publishing issues are at the discretion of the Library].

Procedural—how to satisfy demand

A question was asked about what to do about compliance with ADA laws that demand that all streamed content be made available with cc. There is some frustration with that problem, partially because the process can be expensive. Most libraries are captioning on demand—once there is a student that needs the service, the professor alerts the disability office and/or the library. This is basically the approach of the Temple University Libraries, the UCLA libraries (according to correspondence) and our own much smaller library at MC3. The obvious problem is timing. One idea was to figure out what films are most viewed by students and put these at the front of the queue. [See also University of Maryland’s request forms below]

More from Carleton’s slides below:
Further Notes for Captioning a Film
Format: DVD with Handbrake
Need latest libdvdcss.pkg installed
Output in file containers: .MP4(.M4V) and .MKV
Subtitles (VobSub, Closed Captions CEA-608, SSA, SRT)

Notes on regions and copy protection

libdvdcss(or libdvdcss2)

•Open source software library for accessing and unscrambling DVDs encrypted with Content Scramble System (CSS).

•Used byVLC media playerand otherDVD playerpackages.

•Read and decrypt DVDs of different regions

•Fits DMCA exemptions and fair use under many circumstances; software not yet challenged