A transcriber will listen to an interview and type it word for word. Transcripts often have a heading format giving some general information, including interview name, interviewer, date of interview, location of interview, and a catalog number.
When seeing a transcript for the first time, people often think it takes the format of a play script. Notice in the example below the back-and-forth of the conversation, the exact representation of the words spoken, and the (??) symbol for the place where the transcriber was not able to understand the word.
Note: Transcription can be a time-consuming and costly process. According to Davis, Back and MacLean (1977), for every one hour of interview you are looking at eight hours of transcription work.
Example of transcript format:
Interview with John Doe
December 25, 2009
Interviewer: Rudolph the Reindeer
RR: Good morning. Today is December 25, 2009 and we arre in the home of John Doe, located just south of the North Pole. How are you today, John?
JD: Great! I am so glad you are here.
RR: Now uh- let's start with (??) let's start with the easy questions where were you born?
JD: Well, I was born in Illinois in a small town just west of Springfield, Il.
RR: Is that where all of your family is from?
The Audit-Edit prpocess consists of an audit editor sitting with the transcript and the audio or video interview and reading along while the media plays. An Audit Editor will make changes in format, spelling, crutch words, and inaudible speech to try to strike a balance between ease of reading and integrity of the interview. Remember these will be thought of a primary source documents, so make them as close a representation of the interview as possible.
Format: Make sure all margins match, spacing is correct, and that the font is appropriate.
Spelling: The Audit Editor is considered the subject matter expert and should determine if proper names and places are spelled correctly. The Internet can be a great resource in this activity.
Crutch words:Crutch words can be stutters or words repeated often. At this point in the processing these words can be reduced for easier reading but should not be entirely eliminated. Try to strike a balance between ease of reading and the integrity of the interview.
Inaudible speech (??): Attempt to understand what is being said. Sometimes research on the subject can help. After four or five attempts, move on.
After the Audit/Edit the document should be a good representation of the interview and something that you would be happy to pass on to the interviewee to read.
This part of the transciption processing goes back to the interviewee to give him or her the opportunity to review the transcript, add clarifications, and correct spellings. Send the transcript with a letter thanking the interviewee and instructing him or her to review the transcript for spelling errors and clarifications. The interviewee can make these corrections directly on the paper transcript and return it to you. For ease of future processing, you may ask the interviewee to list page numbers of corrections or circle numbers from a provided chart. This will ensure that you find all the changes requested.
This is the last time anyone will have an opportunity to make corrections to the transcript before it is made public. The Final Editor should check for and implement formatting, requested corrections, and places of inaudible speech.
Formatting: The transcript should be formatted appropriately with a standard space between the speaker's name and the beginning of the text. The header should be in the appropriate format with the catalog number in the top right of each page.
Requested corrections: Look at the self-edit form and you will see where the interviewee has circled the pages where he or she feels that corrections should take place. While the interviewee's opinion counts, we want the transcript to find a balance between an accurate representation of the audio, a readable document, and requests made by the interviewee. We want to delete/remove as little as possible.
Corrections that add additional information are really important and should be inserted in the text in the following format: [insertion].
Requests to delete stuttering and crutches can be accommodated down to a single crutch or stutter so that it represents the audio (i.e. Uh, Uh, Uh, Uh, I had a farm in 2009 changed to Uh, I had a farm in 2009).
Requests to remove portions of the text because of sensitive content should be handled by the Project Coordinator. If a deletion occurs, a note of redacted information will be placed in the text with brackets.
Requests for punctuation change: Read the area in question and listen to the audio at the same time. The form of punctuation that best represents the audio should be used.
Requests for change in spelling: Accommodate all spelling changes that need to be made.
Inaudible words: Skim Through the text for any places with ____(??). All of these areas need to be resolved. If there is a portion that cannot be resolved because of audio quality, replace it with (inaudible speech) in the text.