v1.0.1 - Bugfix

fastText language detection did not work well if the text was in UPPERCASE.

Adding lower() to the string before the language identification step significantly improved the performance.