TL;DR: If we could have even just 4-6 active 3D audio sources on avatars, it would open a giant world of possibilities in creation. The details for those who are curious: I understand the potential reasons behind having a limit of 3 audio sources. Audio crashers were a real problem way back when; but that's just it, way back when. VRChat has a lot of adjustments and safety features for audio sources currently, and when it comes to avatar creation, 3 sources is very limiting for complicated audio setups. Here's an example from my personal experience: For the creation of avatars for combat roleplay, I use UdonVoiceUtils in my worlds to increase the maximum avatar audio distance. The weapons on my avatars compliment this by having close, near and far audio. This helps convey information very well for the people roleplaying, as if there's a gunshot, they can hear it from across the map and it will sound correctly spatialized. However, this causes an issue as without the ability to have audio effects on avatars or advanced curves, the only way I can do this is by spawning 2-3 sources at a time. This means that firing a weapon will cut out other audio, such as foley when operating the weapon (reloading and such.) This also means avatars with contact based footsteps will very frequently have their audio culled in favor of either the weapon or the footsteps. Now, these are not HUGE issues as the gunfire gets priority generally, and that's what's most important to communicate to the players in combat. However, with the new VRC Raycast system, the concept of having audio feedback on hitting the environment and players has been in my mind, however the only way I can do that is by reducing the amount of active audio sources on the weapon... however that means I'll have to sacrifice one of the layers of the firing audio, which consequentially may make the feeling of combat less enjoyable and / or reduce the amount of information available to the players in a given moment. Granted, I know my specific case is a niche utility, but I feel like for many other creators, having even a marginal increase to the amount of allowed active audio sources could open a world of possibility. I've seen discussion on this canny for things like DJ avatars and complex animations, and I fully agree with that. If we had, say, 4-6 allowed active 3D audio sources, it would really open up a huge range of possibilities without going too far and risking audio crashers / unoptimized audio setups. Especially considering the relatively recent implementation of a new audio engine, I feel like it should be well within the realm of possibility to give creators more breathing room with audio. I know that for me personally, it would make creating roleplay avatars significantly easier and allow me to improve the fidelity of them significantly. This doesn't have to impact performance much either, especially if there is a hard limit placed on the amount of audio sources on an avatar. You really don't need more than 32 if your project is contained and you know how to use them effectively, and that's being incredibly generous to be honest. If there is a deeper technical limitation behind this, or it doesn't operate the way I say, please let me know. I want this for the benefit of the community and as such if there is a reason this isn't possible then I want to be able to at least educate people on that reason. Please bring attention to this as well, I don't expect this to get a ton of traction or even for this change to come ever, but it would be awesome if it saw a little focus.
