we often need to know the total number of words in a given corpus, or collection of eaf files. it would be great to implement an easy way to do this in elan, maybe within “search in multiple eafs”, as it is also there, that a corpus(-domain) is defined.
Hmm, I think there are two options that come close to what you want.
The multiple file export “List of Words” (File->Export Multiple Files As->List of Words), with the option “count occurrences” selected, creates a two column text file (tab-delimited) with all unique words and the number of occurrences of each word. Opened in a spreadsheet application the number of rows represents the number of unique words, the sum of the numeric values in the second column should be the total number of words.
Using a regular expression (e.g. for word boundaries) in The “Structured Search in Multiple eaf’s” could also be used to render the word count, I believe.
-Han
I would like to bring up the word-count issue again because, in various projects I’m involved in (e.g. http://doreco.info/), we still dearly miss an easy and practical way to get information on the number of transcribed words for a given corpus (= collection of eaf files). We currently do this by exporting transcription tiers of individual EAF files (usually multi-word “utterance” tiers), then opening them in a word processor and do the word count there, and then adding up these counts for sets of eaf files. But that’s clearly not ideal. So I’d like to stress again, how much we - and we think this is quite common - would benefit from an easy way to count words within ELAN, i.e. counting the number of words (delimited by white spaces) in a given set of tiers (or tier types) across a given set of EAF files, maybe a a variant of the “List of Words” - function.
Thanks for reminding, this would indeed be a useful addition. We’ll add this to the next release, probably either as an extension to the List of Words function or as a variant thereof. I’m sorry this was one of those issues that got lost in the list of things to do!