With inputenc
errors that don't give more details about the offending character (in contrast to, say How to find this unicode character U+202F in your LaTeX documents?, Biblatex can't read bib-entry) it is important to isolate the problematic characters (the problematic line in your .tex
file, the problematic entry in your .bib
file). Even more so because the character that causes the mayhem may be hard to spot with the naked eye. It might be a non-standard space, dash or punctuation character that is virtually indistinguishable from the good character in your editor's font. But of course it could be something easier to spot, a accented character, one with umlauts or other diacritics, a non-ASCII char, ....
One good method to isolate the problem is binary search, where you delete half of your document - make sure not to introduce errors doing that, don't tear environments apart, don't split braced groups - and see if the error persist on recompilation.
You might have to run LaTeX several times until the error disappears if it lives in the temporary files. This is especially important for the bibliography, where you have to run at least LaTeX, Biber, LaTeX to be able to determine whether the error is gone for good. To be really sure you can delete the temporary files (.aux
, .bcf
, .bbl
, ...) between compilcations.