Quick Start

Your first project, step by step

Follow one complete path through Exempla: create a project, add a few texts, check the import, find a word in context, and preserve the results you need.

10 min
Before you begin

Start small, then scale up

Two or three plain-text transcripts are enough for this walkthrough. A small practice project makes the structure easy to see before you import a full corpus.

A project nameFor example, “Sociolinguistic interview 01”.
A few .txt or .docx filesExempla accepts UTF-8 plain text and Microsoft Word documents.
Speaker codes, if neededUse a consistent pattern in every transcript.

Show Exempla what is speech and what is not

A new user should prepare the transcript structure before importing. Put the speaker ID that begins each turn in square brackets. Put comments or transcription notes that you do not want included in word counts or Concordance results in parentheses.

Sociolinguistic interview 01.txtPlain text · UTF-8
[INT]Where did you grow up?
[MARINA]I grew up here in Ottawa. My whole family lived in the region.
[INT]How was it growing up in Ottawa?
[MARINA]Square brackets identify the speaker and begin the turn.
(transcription comment)Parentheses mark comments that should not be counted as spoken words.

Illustration

Keep IDs consistent

Use the same code whenever the same person speaks, including across different conversations and transcripts. Exempla keeps those occurrences together. To protect the reliability of your data, it asks you to review a speaker code whenever an imported transcript uses a code that already exists in the corpus.

Trace every occurrence

Every Concordance result identifies the text file, the speaker code, and the line in the transcription. These references let you return directly to the source passage.

1
Create a home for the corpus

Create a new project

Choose New Project from the Exempla menu. Use a name that identifies the collection rather than the individual search you plan to run.

New Project
Sociolinguistic interview 01

Your project will keep the imported texts, speakers, Clipboard, and project settings together.

Illustration

You are ready when the new project name appears in the Exempla window.
2
Build the corpus

Import your text files

Choose Add Text Files… and select your files in the system file picker. You can also add a file from Import or drag and drop it into Exempla. File names remain visible in the project so you can trace results back to their sources.

3
Confirm the structure

Review formatting and speakers

Exempla first checks whether speaker codes have matching square brackets and comments have matching parentheses. If it finds a possible formatting problem, it shows the file and location so you can review and fix it, import anyway, or cancel. Your original source file is never changed.

Possible Formatting Errors
!
2 possible formatting errors

Sociolinguistic interview 02.docx

Word paragraph/break 14: opening [ has no closing ]

Illustration

After the formatting check, Exempla reviews speaker identity. If an incoming speaker code already exists, it asks whether that code identifies the same person. The file name is shown so the decision has context.

Speaker Code Already in Corpus
?
The speaker code “NARRATOR” appears in the incoming text “Sociolinguistic interview 02” and is already used in “Sociolinguistic interview 01”.

Does “NARRATOR” identify the same person in both texts?

Illustration

Same Person

Combines occurrences under the existing speaker.

Different Person

Lets you assign a distinct code before the import continues.

5
Preserve useful evidence

Select, collect, or export

Select the rows you want to revisit and add them to the Exempla Clipboard. The Clipboard is a project workspace that gathers selected Concordance results from different searches into one table. Export a CSV when you need selected results or a complete result set outside the application.

Exporting CSV
Exporting 7,540 of 12,000 rows…62.8%

Illustration

First project complete You have imported, checked, searched, and preserved evidence from a corpus.

Where to go next