This is a demonstration of a language guesser, as proposed in Cavnar, Trenkle, N-Gram-Based Text Categorization. It's implemented in Perl. You can get the Perl script under certain copyright restrictions here. For free! No commercial version available!