comparing text program.for Kannada script (utf-8)

[74yrs old]

Hi,
Will you help me  with small program "compare text between input text with output text and highlighted misspelling in the output text and total percentage mistakes occurred;
for example:
*Input text*    This is testing the software compare text = 35 chars
*output txt    * This is te*x*tin*k* the so*a*ftw*o*re c*e*mpare t*a*xt = 6  mistakes ( 2.1% mistakes).

It is clarified that I want to locate or highlight the number of spelling mistakes in the output text when comparing with input text and not for  corrections or editing the incorrect spelling required..
One way it is nothing but  statical report of performance of OCR output to know how many spelling  mistakes occurred .in the output  text.
Attached Kannada texts (utf-8) for needful.
In fact I had downloaded windiff - which is not helpful nor to my expectation. As such I approach you for help.
I shall be thankful to you, if you help me.
With Regards,
-sriranga(75yrs old)

kan-input.txt
Description: Text document

kan-output.txt
Description: Text document