non ascii characters are deleted during conversion

[abubelinha]

I have been looking at the provided examples and I think torsimany is not properly converting some encodings.

See this line:
https://github.com/PolBaladas/torsimany/blob/master/Examples/loklak.json#L217
"text": "Инвестиционная идея: XEROX: http://youtu.be/_lD1AuNzSJY?a на @YouTube",

I can't find those cyrillic characters in the converted markdown example. They are replaced by nothing:
https://github.com/PolBaladas/torsimany/blob/master/Examples/loklak.markdown?plain=1#L202
* text: : XEROX: http://youtu.be/_lD1AuNzSJY?a @YouTube

Thanks
@abubelinha