One of my final modules for my Game and Interactive Audio degree is called Interactive Mixing and Delivering. The brief is to take a project we have worked on that uses middleware to implement audio into Unreal and improve the overall mix using interactive techniques. I believe the main aim of this module is to demonstrate that I am aware of mixing techniques and can use middleware effectively to mix a game. The most recent submitted project that fits this brief would be my redesign for the game Seight that can be found in this blog, go check it out if you haven't!
To summarise, I used the middleware Wwise to implement the audio for this project into the game. I believe that in this project I successfully created the overall goals and targets of the sound design I set out for whilst also improving my own understanding of the middleware and it's capabilities. One issue with the project I know for sure is that the overall mixing could definitely be better, with some sound being far too present than necessary. Once I understood the brief for the module I decided upon using this project as I could already think of ways to make the mixing more interactive. Here is a list of the ideas I thought had potential:
Changes to the mix dependant on the height you are at
A counter for each attempt the player makes that effects an element of the mix
A narrative element - linked to time left
A difference between the gameplay states
Different mix states for playback method (i.e. Headphones, TV, Speaker systems etc)
So far I have managed to set up the various buses I want to start with. This consists of the basic categories such as ambience, music, player sounds and other sounds. Then, when I thought necessary I gave more specific sounds their own buses in case I think they might need to be effected individually. This can be seen on the left of the image below.
As you can see, I have set up a basic mix session in Wwise as well as a basic soundcaster session and have started comparing basic loudness of each element to others. Me and my fellow students have come across issues getting the changes we are making in Wwise to effect the sounds within the associated Unreal project too. But, alas, the easy fix is just to remember to include the correct bus, or any for that matter, into the soundbank you are generating. I assumed the changes would be made without needing to but hey, you have to remember I am still learning. Wwise definitely has a way of making me feel embarrassed for forgetting very simple steps. Just another lesson in the important steps to remember, I am unlikely to forget to do it now.
As can be seen, I also have set up some states previously which I can use for the previously listed ideas. I know I can manage to make and implement the ideas I have into this project but I am taking each one step by step. Some mixing tools I have been made aware of include the use of game states as well as 'Auto-ducking'. The latter is a useful and easy to set up tool for ducking one sound when another is triggered. The issue with it is that the controls available do not allow for the ducking to be very smooth as it waits for the end of a track in order to put the volume back up. In FMOD, an easy way to set up something similar is through the use of a Sidechain, not perfect for every situation, but they can help some sounds stand out from the mix. Setting up a Sidechain in Wwise is not as simple but I had a go at making one using a parameter and the effect can be seen in the video below.
As can be heard, the volume of the rocks falling sound makes the volume of the music decrease at the same time, allowing the rocks to stand out more from the mix. Even though the effect is fairly subtle and simple, I am happy that I now know how to set one up and use it in the future since it is more smooth than using the auto-ducking feature.
To progress this project further I need to set up the features I listed at the beginning. For most of them, I can use States to effect the mix and switch those within game. The narrative element I thought of could involve me recording some voice lines that are triggered when the time reaches a certain point. This would allow me to have an extra element in the mix to effect the main categories. Alternatively I could rely on the changes to the mix based on progression to portray narrative without actual dialogue or have a mix of both. I also need to research other useful mixing techniques for similar games, especially tips for mixing within Wwise that I can replicate in this project. So let me know if you know of any useful tips or links to check out and thank you for reading through this blog post.
Comments