Yake ignores preprocessing

[processo]

Describe the bug
Yake keyword extraction outputs capitalized words after lowercasing them in preprocessing.

To Reproduce

1. Corpus with book-excerpts.tab
2. Preprocess Text with lowercase
3. Extract Keywords with YAKE!
4. unlowercased names appear with NA on TF-IDF and Rake columns

Expected behavior
Everything should be lowercase after preprocessing.

Orange version:
3.39.0

Text add-on version:
1.16.3

Operating system:
Windows 10