This presentation will describe some of the techniques that will shape the future of sound design for video games. Two very important technologies will be introduced in particular: Procedural Audio (real-time synthesis) and Audio Features Extraction (audio analysis).

Procedural Audio:

One of the main tasks of a sound designer working in the game industry is to transform static sounds into dynamic ones. Current solutions such as randomization, scripting and modular patching will be presented as well as their shortcomings (not dynamic enough, time-consuming etc...). Then procedural audio will be introduced. It will be shown that it has multiple advantages over these methods (truly dynamic, low memory footprint etc…) and examples of real-time synthesis of air, water, impacts and creature sounds will be demonstrated.

Audio Analysis:

By knowing more about the audio assets we are manipulating, we can offer adapted visualization modes and pertinent processing options to the sound designer. Audio analysis can therefore be used to create smarter and more intuitive sound tools. In addition, we will show how it can be used to improve audio engines by supporting dynamic mixing and audio shaders. Audio analysis is also useful to define new audio-related gameplay.

Combining them together:

Finally it will be demonstrated that by combining these two technologies in the future we can
build even better tools that allow sound artists to design sound effects in much more creative and productive ways. An example of a tool will be presented in which a sound designer can analyze existing samples and transform them into dynamic sound models. Examples of design of physics-related sounds (impacts and contacts) will be shown.