WordHoard is a free Java application, utilizing Northwestern University's Java Web Start plugin, for the study of large texts or transcribed speech. It enables the annotation or tagging of a text by applying morphological, lexical, prosodic and narrotological criteria, and can be applied either to the corpora built into the tool or to a user-provided text. Its functions also include concordance, word frequency, lemmatization and text comparison. It is designed to run on Windows, Mac OS X and Linux/UNIX systems alike, and includes an auto-update feature to ensure all users are running the most current version without the need for users to monitor for updates. Though it utilizes web content, the application runs from a cache stored on one's hard drive.

On launch, WordHoard prompts users to choose from one of two WordHoard sites, each containing a different base corpus. The first, the Main Public Site, contains Early Greek Epics and the works of Shakespeare, Spenser and Chaucer and is open to all users. The second, the Early Modern English Drama Site, contains approximately 320 Early Modern English plays (including Shakespeare's plays), but is only available to institutions subscribing to the Text Creation Partnership (TCP). For users interested in these corpora, WordHoard offers rich metadata for each text in keeping with their canonical status, and this metadata can be leveraged in various ways by the tool to enhance users' investigation of them. This metadata notably includes narratological and prosodic information, and metrical parsing of the Early Greek epics.

WordHoard is easy to use, with well-labelled menus and options for investigating a text. This is well-supplemented by a clear, comprehensive user guide. The guide is indexed by topic and easily accessible from the main page, including install, troubleshooting and how to use each of its features. Of interest to advanced users, it also details such features as the application's built-in shortcuts and how to adjust WordHoard's look and feel to better suit UNIX users. Custom queries can also be saved for later use. New users can expect to begin meaningful analysis of a text within minutes of downloading the tool if they use the provided corpora.

Using WordHoard with custom corpora may be challenging for some users. Users aiming to add their own texts will need to familiarize themselves with the application's directory structure and its raw data files before proceeding. Texts must be in XML format with UTF-8 encoding, and additions must be made to the application's raw data files in order for WordHoard to recognize and process the new corpora. Full instructions are included in the user guide.

Linux/UNIX users will need to install a font suitable for displaying text in the Roman and Polytonic Greek alphabets to make the best use of the corpus provided (Windows and Mac OS X include suitable fonts already). The WordHoard development team recommends Gentium, and has provided a link for ease of access.

As a whole, WordHoard is a mature, well-designed text analysis tool, though users seeking to apply this tool to their own corpora may find it overly challenging. It is particularly recommended for users interested in Early Modern English and Greek texts. For users comfortable with the complex corpora addition process, it is also of high value to those seeking further insight into their own texts. However, users looking for a quick, easy way to delve into texts beyond the provided corpora should be advised that WordHoard may be inconveniently complex for their needs.

WordHoard is no longer in development as of 2013. However, the website is still up and the software still available.

WordHoard was mentioned in this research abstract https://dh2017.adho.org/abstracts/408/408.pdf