This is a white paper for a project that harvested audio and transcribed data from podcasts and news broadcasts on the web. Tools were developed to analyze the different uses of prosody (rhythm, stress and intonation) within spoken communication using phonetic analysis and machine learning.