The 300 Languages Project is a special effort to begin the construction of a universal corpus of human language by collecting parallel text and audio in the world's 300 most widely-spoken languages. The resulting collection will contain thousands of volunteer-contributed public domain text documents and audio recordings