Step-by-Step: Converting Your Transcript to an OHD-friendly CSV Format

This beginner-friendly tutorial guides you through converting a standard interview transcript into the structured CSV format used by Oral History as Data. Follow this step-by-step process to prepare your first transcript.

What You’ll Create

By the end of this tutorial, you’ll have:

  • A properly formatted transcript CSV file
  • Content ready for upload to your OHD site
  • A file that can be visualized and searched

Before You Begin

Gather these materials:

  • A raw interview transcript (Word document, text file, etc.)
  • Access to Google Sheets (recommended) or another spreadsheet program
  • 15-20 minutes of time (less for subsequent transcripts)

Tutorial Steps

Step 1: Create Your Spreadsheet

Go to our template spreadsheet and click “Make a copy”.

Step 2: Clean Your Transcript Text

  1. Open your transcript document
    • Open your interview transcript in Word, Google Docs, or another program
    • Select all text (Ctrl+A or Cmd+A) and copy it (Ctrl+C or Cmd+C)
  2. Use the text cleaner tool
    • Paste your text in the top box below
    • Click the “Clean Text” button
    • The cleaned version will appear in the bottom box
 
  1. Copy the cleaned text
    • Click in the bottom box
    • Select all text (Ctrl+A or Cmd+A)
    • Copy it (Ctrl+C or Cmd+C)

Step 3: Structure Your Transcript

  1. Paste text into your spreadsheet
    • If you don’t have any Speaker’s noted in the text, Click in cell B2 (under the “words” column)
    • If you do have Speaker’s noted, click in cell A2 (under the “speaker” column)
    • Paste your cleaned text (Ctrl+V or Cmd+V)
    • The text will appear in a single cell
  2. Separate text by speaker
    • If your text appears in a single cell, with the Speaker’s name starting each cell like so “Speaker:” follow these steps:
      • Select the cell with your text
      • Using the menu at top, click Data → Split text to columns
      • In the separator options, choose “Custom” and enter a colon mark –> “:”
      • Click “Split”
    • Speaker’s should now be separated from the text that they speak!
      • Note: if you have colons in the rest of your text, this will also be split out into new cells; you can paste that content back into the preceding cells, or just replace the current words column with the cleaned transcript you had from before, then do a find/replace function to remove any mention of speakers
  3. If not Speakers are listed, Identify speakers
    • In column A, add the speaker name for each segment of text
    • For example:
      • “Interviewer” for questions
      • “John Smith” for the interviewee’s responses
    • Be consistent with names throughout

Step 4: Add Optional Information

  1. Add timestamps (if you have audio/video)
    • In column D, add timestamps for key segments
    • Format as MM:SS (e.g., 01:45) or HH:MM:SS (e.g., 1:12:30)
    • Match to the corresponding points in your recording

  2. Add topic tags (for visualization and analysis)

    Create your tag vocabulary first:

    • Before tagging segments, create a filters.csv file in your _data folder
    • Include two columns: tag and description
    • Review all your transcripts to identify 10-20 common themes

    Example filters.csv:

    tag,description
education,Educational experiences and schooling
family,Family relationships and history
career,Work and professional life
childhood,Early life experiences
community,Community involvement and local history

    Apply tags to transcript segments:

    • In column C, add relevant topic keywords from your vocabulary
    • Use semicolons between multiple tags (e.g., “education; family; career”)
    • Be selective - focus on the most meaningful content
    • Use exact tag terms from your filters.csv file
    • Keep formatting consistent: lowercase, no special characters

Step 5: Finalize and Save

  1. Review your spreadsheet
    • Ensure each row has a speaker identified
    • Check that text is properly separated (one speaker segment per row)
    • Verify any timestamps or tags are correctly formatted
  2. Save as CSV file
    • Click File → Download → Comma-separated values (.csv)
    • Find the downloaded file (don’t open it!!) where it’s been downloaded
    • Rename your file to match the objectid in your metadata
    • Example: If your interview has objectid “smith_john” in metadata, name the file “smith_john.csv”
  3. Upload to your repository
    • Place the CSV file in the _data/transcripts/ folder
    • For GitHub users:
      • Navigate to the _data/transcripts/ folder in your repository
      • Click “Add file” → “Upload files”
      • Drag your CSV file or use the file selector
      • Commit the changes

Example of Completed Transcript

Here’s how a properly formatted transcript CSV should look (note: this is overly tagged to serve as an example … )

Speaker Words Tags Timestamp
Interviewer What was your first teaching job? career; education 00:15
John Smith “I started at Lincoln Elementary in 1972. It was a challenging environment but rewarding.” education; career; 1970s 00:22
Interviewer What subject did you teach? education 01:05
John Smith “I taught sixth grade math and science, though I preferred the science lessons.” education; science; mathematics 01:12
Publishing Your Site Metadata Creation