Drop in an XLIFF file and get the translatable word count, with locked units excluded and repeated segments separated out. Those two exclusions are where a quote quietly inflates, and knowing the numbers before the conversation is worth more than checking them afterwards.
Free · No account · Your files never upload
Version 1.2 or 2.0. Read on your computer and never uploaded, which matters when the file is a client’s unreleased course.
Reads an XLIFF file, in either version 1.2 or 2.0, and counts the translatable words. Locked units are excluded and repeated segments are separated out.
Units marked as not for translation are usually product names, code and identifiers. They are in the file because they have to be, not because anyone should be paid to translate them, and counting them is the commonest way a quote quietly inflates. Repeated segments are the other one: a segment translation memory has already seen is normally discounted, so the unique figure is the one that survives to the invoice.
Version 1.2 and version 2.0 are structurally different, not just renamed. A reader that only understands one returns zero units on the other and looks like it opened an empty file rather than an unsupported one. Both are handled, and a file with no version attribute is identified by its shape.
It counts what is in the file. If the export left something out, the count will be short by exactly that much, which is worth checking against the Course Word Count run on the original source file.