A video content generation device generates video data synchronized with music data based on motion data, representing a motion graph including nodes, edges, and weights, and metadata indicating a synchronization probability per each node between the motion graph and a musical tune. A music data storage unit stores the predetermined amount of music data and their musical characteristics in connection with the predetermined number of beats, in a reproduction order, retrieved from the musical tune. An optimum path search unit searches an optimum path connecting between nodes, each of which is selected per each beat with a high synchronization probability, on the motion graph with motion characteristics matching with musical characteristics based on the predetermined amount of music data. Video data synchronized with music data is generated based on synchronization information for correlating motion data to music data along with the optimum path.