Is there a text expander that only predicts based on the current text?

I have to type a lot of text in a transcription program called InqScribe. I can’t use a discrete text processor like LibreOffice, which has built-in text expander that predicts long words that you’ve already typed within your current document. I want to have something like that that would work with InqScribe but not a global text expander that learns from what you type in Word, on the internet, etc. but just from the document you’re currently working on. Any ideas?