Skip to Main Content

Oral History - Methodologies and Sources

This guide brings together information on oral history methodology as well as a list of select oral history collections and resources at Duke Libraries.

There are countless oral histories out there languishing in dusty piles of cassettes on researcher desks, filling up shoeboxes in closets, and even still sitting in the recorders that captured them, unseen and unheard.  This is because the other significant piece of oral history work, and some would say the harder part of the oral history endeavor, is managing the interview after it’s been recorded.  However, if the groundwork has been done – the interview properly planned and a place for the associated documents to “live” established (See Logistics) – then much of the processing burden has already been negotiated, by simply knowing up front how the interview will be handled after its production.

From oral history to historical record

The first priority following an interview is to back up the recorded audio or video media.  Given the current technology, this means copying the file from a solid state medium like a flash card to another drive, either local, on a network, or in the cloud, and naming it appropriately given the names of the other documents you've generated for the interview.  Following the principle of LOCKSS (archives-speak for Lots Of Copies Keeps Stuff Safe), these measures will help ensure the interview's long-term well-being.  Be aware, however, that any solution -- even the "safer" cloud drive option where you're paying for space -- requires monitoring over time.  If you decided to take a chance with a proprietary audio format that just seemed like a good idea or was your only option, consider converting the file to a WAV audio file or a higher-quality MP3.  Doing this upfront may save you some headaches down the road.

A good, free audio editing program that will help you transcode audio to different formats is Audacity.  Audacity is also great for cleaning up your audio, making trims where necessary, adjusting volume, and applying fade-ins and fade-outs.  It can perform many other audio editing functions as well, but these are the most common ones for oral history, which shouldn't require much further editing unless the interviewee requested specific sections be cut.

Of indexes, transcripts, and captions

Despite the immediacy of the personally-voiced expression of an eyewitness to history, the quality of the textual description often decides which interviews get used in archives and which don't.  Oral historians love audio, researchers want text.

There are three basic kinds of textual description of oral history: subject analysis indexes, transcripts, and captions.  Indexes describe chunks of audio, offering a subject analysis of a part of an interview demarcated by time.  An index entry looks something like this: "00:05:00-00:10:00, participation in Greensboro Sit-ins; entering graduate school at UNC, 1961; Ph.D. work, UMass Amherst, beginning 1965."

Transcripts are texts of what was said by the interviewer and interviewee, often created with a set of rules, which may allow for skipping false starts, "umm's" and the like.  Some transcriptionists are also given leeway to change grammar fairly dramatically.  A transcript therefore may not contain exactly what the interviewee said.  Additionally, because the transcript was long considered the final product of the interview, the narrator was often given the right of editorial approval prior to opening the interview to use, with sometimes questionable results.  Those new to oral history are often surprised to learn of the intentional discrepancies that occur between transcripts and their source audio.  Rather than think of the transcript, then, as a textual backup of the recording, instead it's probably better to think of it as enhanced metadata, a close description of the audio.  Sometimes transcripts will be indexed to show, within the text itself, time elapsed (typically every one or five minutes), to aid researchers interested in accessing the audio.

Captions -- verbatim text transcripts continuously timecoded -- are increasingly common, as: 1) playback software develops towards supporting multiple streams of media, accommodating text searches; and 2) the Americans with Disabilities Act is increasingly enforced in libraries and educational institutions.  While captioning an interview can be costly, and requires a playback interface that can read the captions to be effective, it allows for text-based access to the full interview and, therefore, access to a wide population of potential users.

Tools and vendors

Oral History Metadata Synchronizer (OHMS) - a free indexing tool and online viewer developed by the Louie B. Nunn Center for Oral History, OHMS became publicly available in 2014.  The indexing tool creates an XML file that the viewer uses to link the media and the description.  OHMS was created for both individuals and institutions, and provides robust help for setup, only requiring that the user has access to a web host running PHP 5.3 or greater.

Express Scribe - free transcribing software from NCH, Express Scribe is essentially an audio player that can be easily joined to a USB foot pedal controller, keeping hands free for typing.  Google around for a foot pedal -- they can be had for $25-50.

University of Washington Video Captioning Guide - a handy rundown on how to caption your own video for free.  Bear in mind this will work equally well for audio, you just need to create a video from your audio file (check out services like TunesToTube).

There are a number of vendors who can help with transcription and captioning.  Their costs differ based on whether they use primarily software or humans for transcription, and your tolerance for errors (humans, naturally, cost more but are also more accurate).

Pop Up Archive uses software for transcription and has lots of experience working with libraries and non-profits.  Be prepared for lower accuracy rates but a good price and a good interface for helping edit out the problems.

Rev, like Pop Up, offers a quick turnaround, but is more expensive (they use humans).  Rev also provides a handy interface for editing and can provide all kinds timecoded files depending on your needs.

Audio Transcription Center has a 50-year history in the transcription business and is extremely reliable.  They cost more but their accuracy rate and customer service is exceptional.