Music has long been an integral part of human culture, serving as a means of expression, communication, and entertainment. Melody, the sequence of single notes that forms the “tune” of a song, is a crucial element in music composition. It is the part of a song that listeners tend to remember and hum along to. Over the years, musicians and composers have used their creativity and expertise to craft memorable melodies. However, the process of creating a compelling melody can be both time-consuming and challenging. In recent times, advancements in artificial intelligence (AI) and machine learning (ML) have opened up new possibilities in the realm of music creation. Researchers have explored ways to automate or augment the process of melody generation, aiming to develop algorithms that can learn from existing musical data and create original compositions. This not only has the potential to democratize music creation by enabling non-musicians to generate melodies but also offers a new perspective for professional musicians, serving as a tool for inspiration and experimentation. MATLAB simulation shows that under the condition that the composition requirements are certain, the score production accuracy and melody selection time of the GRU algorithm are better than those of the online composition method.