A software architecture for dynamic enhancement of soundscapes in games

A game soundscape often includes sounds that are triggered by the game logic according to the player's actions and to other real time occurrences. The dynamic nature of such triggering events can lead to a composition that is not necessarily interesting, at a given moment. Following principles from Acoustic Ecology, and specifically the notion of healthy soundscape, we propose a system aiming at the moderation of sounds generated during gameplay in a way that the composition retains its communicational meaningfulness. For instance, the system aims to ensure that the soundscape does not get overcrowded by the superimposition of whatever sounds might be triggered, and that the player can actually ear the relevant stimuli. The main component of the system is a module that implements heuristics that free designers from restating accepted sound design principles, and programmers from embedding such intelligence in the game logic. Thus, designers can focus on expressing their design intents, by using an API to create and characterize the sound sources to be handled by the sound engine. The use of this API also conveys a more readable expression of the game's sound design, easing communication and reuse. Hence, this proposal can be particularly relevant during fast prototyping phases, when it can constitute an expedite way to test and refine creative ideas, while avoiding extensive coding and the typical complexity and cost of existing middleware, which is particular relevant for small budget and sound novice practitioners. We finish by presenting results from a proof of concept implementation, and a game remake for evaluating the proposed system.