When you listen to a foreign language being spoken, it likely sounds like a single stream of sound rather than a series of distinct words. To parse that stream of sound into distinct words, what would you need to do$?$