Automatic Generation of Melodic Accompaniments for Lyrics

Music and speech are two realms predominately species-specific to humans, and many human creative endeavors involve these two modalities. The pairing of music and spoken text can heighten the emotional and cognitive impact of both the complete song being much more compelling than either the lyrics or the accompaniment alone. This work describes a system that is able to automatically generate and evaluate musical accompaniments for a given set of lyrics. It derives the rhythm for the melodic accompaniment from the cadence of the text. Pitches are generated through the use of n-gram models constructed from melodies of songs with a similar style. This system is able to generate pleasing melodies that fit well with the text of the lyrics, often doing so at a level similar to that of human ability.