Transcription

Step 1. Understand our approach to transcription

We will present each item we edit from the Eartha M.M. White Collection in two versions:

  • a transcription that attempts to record the original as it is, within certain limitations, and
  • a partially regularized reading view that makes the document more accessible.

In order to present the first, we will attempt to capture many (but not all) aspects of the original as we transribe.

1.1 What we do and don't reproduce

We will reproduce the following aspects of the original:

  • spelling,
  • word division,
  • punctuation (but we will standardize sequences of multiple periods separating items in event programs, etc., to just five, as explained in Intermediate Encoding),
  • use of capital letters, 
  • hyphenation of words divided across lines,
  • abbreviations,
  • division into paragraphs,
  • line breaks (each line in the original should be a line in your transcription), and
  • page breaks.

Additionally, we will reproduce the following, but you will need to handle these in the next step (Basic Encoding):

  • underlining,
  • italics and bold type (in print documents),
  • superscript, 
  • interlinaeal or marginal insertions,
  • struck text, when legible,
  • indentation of paragraphs, and
  • approximate horizontal spacing of elements, with limitations.

We will not reproduce:

  • font,
  • text size,
  • spaces between initials (we transcribe A. L. Lewis as A.L. Lewis), nor
  • any spacing beyond a single space between words.

In printed booklets and similar documents that have internal page numbers and running headers (text that runs across the top of every page), we omit those in our transcriptions.

1.2 Basic guidelines

Do:

  • type what you see, reproducing everything in the first list above, and
  • ignore everything in the second (for now) and third lists above.

Don't:

  • make any modifications to spelling,
  • correct anything you might regard as an error in the original, 
  • resolve any abbreviations, or
  • do anything that would alter the way the text would sound when read aloud.

1.3 Tools and shortcuts

1.3.1 Can I use AI for transcription?

An example of using Chat-GPT for transcription

Yes. Chat-GPT, for example, is good at transcribing both typewritten/print document and handwritten documents, in English and Spanish. You can upload a single image file, or you're working with a document from the Eartha M.M. White Collection with multiple pages, check to see if there is a value in the column for "Link to item on Digital Commons" on our document spreadsheet. If so, you can download a PDF with all the pages from there. 

Try this prompt:

Make a transcription of this document, preserving the line breaks and keeping all the following as-is spelling, word division, punctuation, use of capital letters,  hyphenation of words divided across lines, abbreviations, division into paragraphs. Indicate where each new page begins. Ignore blank lines and extra spaces between words.

If you're working with a document in Spanish, try this prompt:

Realice una transcripción de este documento, conservando los saltos de línea y manteniendo todo lo siguiente tal como está: ortografía, división de palabras, puntuación, uso de mayúsculas, separación de palabras divididas en líneas, abreviaturas, división en párrafos. Indique dónde comienza cada nueva página. Ignore las líneas en blanco y los espacios adicionales entre palabras.

After you cut and paste the results into your XML file, you will need to adjust the text to match the original document in terms of everything in the first list in 1.1 above.

Note that the accuracy of the output which will vary, depending on the conditions of the original. 

1.3.2 Can I use OCR?

For typewritten and print documents, you can use Optical Character Recognition (OCR), which turns an image into text. Several free OCR services are available online, including one from Adobe. OCR is also now incorporated into many applications that display images. Some OCR programs may now also handle handwritten images.

Some online OCR services will accept image files, and some may require you to upload a PDF. If you're working with a document from the Eartha M.M. White Collection, you may be able to download a PDF from the UNF Digital Commons (check to see if there is a value in the column for "Link to item on Digital Commons" on our document spreadsheet for your item). 

After you cut and paste the OCR results into your XML file, you will need to clean up the text and make sure that it matches the original document in terms of everything in the first list in 1.1 above.

If you need help running an OCR scan of your document, contact Dr. McCarl.

1.3.3 Can I use speech-to-text?

Yes, instead of typing the contents of the document, you could also use to the speech-to-text functionality in Word, Google Docs, or any other program. However, you will need to clean up the results to make them match the original document in terms of everything in the first list above.

1.3.4 Why is my document already transcribed?

This section applies to all students in Summer 2026. Everyone is working with an initial document that was transcribed by Tracey Britton in the UNF Library.

In the case of many handwritten documents, the UNF's Special Collections and University Archives has provided transcriptions. If this is the case for your document, the transcription will already appear in your XML file. The name of the person in Special Collections and University Archives who carried out the transcription should already appear in the TEI Header (make sure you do not remove it).

Note that those transcriptions do not necessarily follow the guidelines established in 1.1 above. You will need to proofread the transcription against the original, adjust it to match our project criteria, and fill in any blanks. Make sure, in particular, that the line breaks in the transcription provided match the line breaks in the original document.

1.4 Transcribing and encoding simultaneously

In the next step (Basic Encoding), you will learn to mark up your transcription in TEI-XML. You are welcome to transcribe your entire document now, before you move on, and if you're using one of the tools/shortcuts described above, that's probably the best approach. However, if you are transcribing manually, you may find that transcribing and doing basic encoding at the same time is more convenient.

Step 2. Transcribe the first 4-5 lines of your document

In Summer 2026, your initial document is already transcribed. All you need to do here is make sure that the first five lines match the original document.

Working from the document images that are available when you view your document (see the corresponding URL on our spreadsheet), transcribe the first 4-5 lines of your text. Place that material between <p> and </p> where currently you see "Delete this text and transcribe here."

If your file already has a transcription in it, your task here is to proof it and make it match our project criteria (see 1.3.3 above).

Make sure that you don't change anything in the file other than adding your transcription between <p> and </p>.

In other words, before you begin, you will see this:

<p>Delete this text and transcribe here.</p>

When you're done, you should have something like this:

<p>This is the first line of the document
and this is the second line
and this is the third line
and this is the fourth line.</p>

You may want to right/control + click on the image and save it to your computer. If you do so, you can then open it in whatever you normally use to view images. You could also open the image on your phone or another device.

Typcally this work is easiest when you have two screens: one where you can view the document image and another where you can type. This could be your phone and a computer, or any other combination that works for you. 

If there are things you can't read or complexities you don't know how to handle, don't worry about them for now. Just do what you can, leaving blanks as needed.

Step 3. Place your file in our OneDrive folder

Once you have transcribed the first 4-5 lines, save your file and then put a copy into this OneDrive folder of files to transfer. 

Step 4. Understand our file transfer process

Barring unforeseen circumstances, I will transfer everything in the OneDrive folder to the server at least twice a day, Monday-Friday. I will attempt to do one transfer by 9 AM in the morning EDT and again by 8 PM EDT. I will transfer files at other times as I am available.

Each time after I do a transfer, I will take everything from the OneDrive folder and archive it in this OneDrive folder of files already transfered. I will create a subfolder each time labeled with the date and approximate time.

I will also make a post to this "File transfer log" discussion forum announcing that the transfer is complete. If you would like to receive a notification each time I do a transfer, you can subscribe to that discussion thread (make sure your Canvas notification settings are configured to send you a messag when new discussion replies are posted).

Otherwise, you can check the OneDrive folders. If your file is no longer in the folder of files to transfer, I have moved it to the server. You can double check by looking in the folder of files already transferred, review my posts on the "File transfer log" discussion forum, or just confirm that your most recent changes are reflected when you look at the file online by following the URL on our document spreadsheet.

Step 5. Review your modified file

Once I have copied your file to the server, you will be able to see your changes.

Please note the following:

  • If you already have the page open, you will need to reload it first to see the changes.
  • If you don't have it open, remember that you can get to it by following the corresponding URL on our document spreadsheet.
  • If you don't see your changes when you load or re-load the page, you may need to clear your browser cache and then try again.

If everything has been successful so far, you will see two changes:

  1. You should see your name at the end of the following sentence at the bottom of the page (Joe Smith is a placeholder here): "Published as part of the project Editing the Eartha M.M. White Collection with the collaboration of Joe Smith." (Prior to the change you made in step 5 of Setup, it read as follows: "Published as part of the project Editing the Eartha M.M. White Collection with the collaboration of Your name here.") 

    If that isn't the case, double-check to make sure you did Setup, step 6.

    If the file was already transcribed by someone in Special Collections and University Archives, that person's name will also appear in this sentence.

  2. You should see the text that you transcribed. It won't look pretty and it will all flow together, but as long as you see your work there, that's all we need for now.

    If that isn't the case, repeat steps 2 and 3 above, as necessary.
    Transcription