Procedural Synthesis of Gunshot Sounds Based on Physically Motivated Models

Generation of content for games is one of the major bottlenecks in terms of the effort required and the resources to be committed. A typical AAA game contains tens of thousands of sound files as audio assets, which include spoken dialogue as well as sound effects. Procedural content generation (PCG) provides a cost-effective alternative to recording these sounds in the studio or in the field. While some sound effects can be recorded with comparatively easy, given the necessary time, effort, and resources, some others such as gunshot sounds are not easy to record. Since many games and simulations incorporate firearms, parametric sound synthesis, which is essentially a PCG technique can be used to alleviate the need to record gunshot sounds. This chapter describes a physically motivated parametric gunshot sound synthesis model. The model is based on a deconstruction of the gunshot sound event into its constituent parts and uses parameters such as the barrel length, bullet type, and muzzle velocity to synthesise the sounds of different firearms. A subjective evaluation, which investigates the perceptual relevance of the proposed model, is also presented.