Creating adaptive music for video games

Knowledge Base | Document

Sound and music design for video games has always been a complex discipline, for a number of reasons. On the one hand, it seeks to replicate very high quality standards, because of the influence of other mass media – especially film and television; on the other hand, the extent and low linearity of today’s interactive experiences make designing adaptive sound with a film perspective a titanic task, difficult for most development teams to address.

Such a system is capable of automatically managing a set of musical fragments designed by a composer

In order to remedy this situation, a group of Narratech researchers has designed an adaptive music system for video games based on emotion recognition, recently presented at the International Advances in Computer Entertainment Technology (ACE) congress. Such a system is able to automatically manage a set of musical fragments designed by a composer, in order to achieve a soundtrack that changes according to the decisions made by each player. Since the method described is based on text analysis, its possible applications on narrative-type video games are very broad. In addition, the research team, composed of Manuel López, Nahum Alvarez and Federico Peinado, is currently working on an extension of the aforementioned system that would add the ability to guide the player on the right path in three-dimensional environments, also through the management of the soundtrack.


“Towards an Emotion-Driven Adaptive System for Video Game Music” is the most recent work done by researchers Manuel López Ibáñez, Nahum Alvarez and Federico Peinado and has been accepted for presentation at the international congress ACE (Advances in Computer Entertainment Technology) 2017.

The article proposes a software architecture for an adaptive sound system, designed in order to improve the listening experience in narrative video games taking into account the emotional response of the user when choosing the music that is played in each playable sequence. This architecture would also allow to compose in real time new melodies that adapt to each situation.

If you are interested in learning more about this topic you can access the full text here