Text Analysis Tools

Voyant is a web-based text reading and analysis environment that enables users to read and explore a corpus of text using a multi-panel interface. Created by Stefan Sinclair of McGill University and Geoffrey Rockwell of the University of Alberta for digital humanities students and scholars, the general public, and those new to digital humanities, Voyant lowers the barrier of entry to using text analysis in scholarly research. Voyant's default tools include a word cloud; a reader to read the documents of the corpus; a line graph that shows the distribution of a word's occurrence across the corpus; a textual overview of the corpus; and keywords in context. Several other tools are also available for users. Voyant is best to use when you have a text-heavy corpus and when you want to explore that corpus using a variety of tools. The interactive visualizations created in Voyant are exportable and can be used to add functionality to online projects, collections, blogs, websites, essays, and more.

• Voyant documentation 
• Voyant tutorial 
• Voyant screencasts

1. Upload a corpus on the Voyant homepage. You can open an existing corpus; type or paste text into the text box; or upload files from your computer. Click reveal when done.

2. You will be directed to the skin, which is the default arrangement of tools. These are the tools that are in the skin:

• Cirrus: Word cloud that shows the most frequent terms of the corpus
• Reader: Fetches segments of text from the corpus as you scroll through it 
• Trends: Distribution graph showing terms across the corpus
• Summary: Brief overview of the corpus
• Contexts: Concordance that shows each occurrence of a keyword with the surrounding context

You are able to change which tools you see in each pane. See Voyant's list of tools for additional functionality.

3. Export your data and visualization. Export options vary by tool. To export, click the export button (square with arrow) in the top right hand of each pane. You will be able to export the view, which includes an HTML snippet to embed the view in a web page, or the data.