Captioning James Farmer’s Reflections Lectures

Written by Francesca Maisano ’21, Digital Archiving Lab Student Aide.

In preparation for UMW’s Centennial Celebration of James Farmer, Farmer Legacy 2020: A Centennial Celebration and Commitment to Action, the Special Collections & University Archives Department, specifically the Digital Archiving Lab, was tasked with captioning a series of filmed class lectures of James Farmer. These date from 1987, when Dr. Farmer was Commonwealth Professor in History at what was then Mary Washington College. Captioning the videos will make the videos accessible for those who have hearing impairments. Transcripts are also very important to include in audiovisual collections and had been previously created by Laura Donahue ’12, Michelle Martz ’12, Kelsey Matthews ’13, and Caitlin Murphy ’12, students in Dr. Jeffrey McClurken’s 2012 Adventures in Digital History class for their website on James Farmer’s Reflections Lectures.

The basic captioning process went as follows:

First, the videos were uploaded to YouTube. YouTube was chosen as it automatically captions videos, so the words and timings would already be there. However, you have to use Creator Studio Classic, not YouTube Studio Beta, as the latter does not do captioning. This took about 10-15 minutes, as it includes both the time for uploading and processing.

Screenshot of YouTube showing a video is being uploaded, as referenced in Step 1.

Videos can be uploaded to YouTube for automatic captioning.

Then, the automated captions were edited. YouTube automated captions can often not be particularly accurate. However, as James Farmer’s voice was so clear and he spoke relatively slowly, the captions were more accurate. Most of the editing was to add punctuation, which automated captions do not have, and check for spelling mistakes, mostly of names. For example, YouTube does not recognize that Fredericksburg is an actual name, so it would misspell it in the same way in every video (the automated captions may have been wrong, but at least they were consistent!). Another name that YouTube did not recognize was CORE, Farmer’s Civil Rights organization, which it would often misspell as “core,” all lowercase. It also did not recognize another frequently talked about organization: the Civil Rights organization, SNCC. I found it easier to edit wording and adjust timing in YouTube than in Adobe Premiere Pro, so most of the editing was done at this stage. While the times varied between videos, as some were a little shorter than others, and some videos’ captions were more correct than others, it would often take around an hour plus to complete the work.

Screenshot of YouTube in which captions of a video are being edited, as referenced in Step 2.

In YouTube, you can edit automatic captions through Creator Studio Classic.

Once the captions were done, they were downloaded then uploaded with the .mpg clip into Adobe Premiere Pro. The captions were lined up and the wording was checked to make sure nothing had happened between YouTube and Adobe Premiere. One common issue was that instead of quotation marks in the captions there would be stars. Another common issue was that text would stretch nicely across the YouTube video, but on the Adobe video the text would run off the video, the line of text being too long, and then the text would be cut off.

Both of these issues required scanning through the text and editing any mistakes in the captions and adjusting the length of the lines of the text if they were too long, adding another 5 to 20 minutes of time to fully complete the work.

Screenshot of Adobe Premiere Pro in which captions of a video are being edited, as referenced in Step 3.

In Adobe Premiere Pro, captions can be edited after they’ve been inserted into the project.

Finally, the videos were exported with the captions from Adobe Premiere Pro, which embeds the captions into the videos. This took between 3-4 minutes per video.

Screenshot of Adobe Premiere Pro depicting a video being exported as referenced in Step 4.

After uploading and editing videos in Adobe Premiere Pro, the videos can be exported into a single, open-captioned video.

Overall, captioning the videos was both an enjoyable and educational project. While editing, there were times I would listen to a sentence or two over and over again to make sure the wording was correct. James Farmer really was a fantastic orator! I also learned a great deal about the Civil Rights Movement. Growing up what I learned about the Civil Rights Movement was surface-level and never for very long. These videos allow for people to learn and understand the history, events, and people of the Civil Rights Movement from someone who was involved in and personally led the movement. Listening to Farmer made it more personal and impactful to me, as it was not just words in a textbook but a person, with his memories and emotions, talking about his experiences. It was not always pleasant to listen to, as he would talk about discrimination, violent racism, and even death, but I always left more knowledgeable on the subject and ruminating on the past, present, and future.

Both captioned and uncaptioned videos along with the transcripts are now available on Simpson Library’s Digital Collections platform. We would also like to give special thanks to the many colleagues at UMW and other institutions who gave wonderful advice on this process!

November 26, 2019