The algorithm itself is language independent. To use it with Spanish you need a Spanish frequency dictionary. The Google Books n-gram data also includes Spanish word frequencies.