Я использовал это:
// Replaces commonly-used Windows 1252 encoded chars that do not exist in ASCII or ISO-8859-1 with ISO-8859-1 cognates.
var replaceWordChars = function(text) {
var s = text;
// smart single quotes and apostrophe
s = s.replace(/[\u2018|\u2019|\u201A]/g, "\'");
// smart double quotes
s = s.replace(/[\u201C|\u201D|\u201E]/g, "\"");
// ellipsis
s = s.replace(/\u2026/g, "...");
// dashes
s = s.replace(/[\u2013|\u2014]/g, "-");
// circumflex
s = s.replace(/\u02C6/g, "^");
// open angle bracket
s = s.replace(/\u2039/g, "<");
// close angle bracket
s = s.replace(/\u203A/g, ">");
// spaces
s = s.replace(/[\u02DC|\u00A0]/g, " ");
return s;
}
//Use like:
var newText = replaceWordChars(textToCheck);
С http://www.andornot.com/blog/post/Replace-MS-Word-special-characters-in-javascript-and-C.aspx