I am attracted to The FUSE (The Finnish Upper Secondary School Corpus of Spoken English) because it is rare to find CEFR level spoken corpus texts.
What I instantly find from the B2 students below is that the complexity checker is not calibrated to spoken texts. Firstly, sentences are broken up with various codes which interrupt the location of structures. I personally don’t have the time to manually go through and edit the data so that it reflects a real level. The complexity checker can still locate some grammar points and vocabulary items though.
