Hello creators! 
We’re excited to announce Acoustic Simulation as a new Studio beta feature! With this feature enabled, audio playback will automatically adapt to its environment. Specifically, any 3D audio emitted using the newer Audio Wiring API (including audio assets, voice chat, and/or character sounds) will now behave similarly to the real world. You’ll hear these sounds being muffled by structures, as well as bending around and reflecting off of parts and terrain.
The best part? This is all automatic—no scripting is required. Sculpt a large stone cave that echoes emitted sounds within and throughout. Create a room that blocks noise when you close the door. Make a winding hallway where users can hear foes from around the corner.
Keep in mind, this will currently only work within Studio and not published places. This will be able to be used in published experiences in the near future!
How do I use it?
You can find a toggle to enable the feature under File → Beta Features → Acoustic Simulation in Roblox Studio.
With Acoustic Simulation, you’ll have access to the following APIs:
-
AudioEmitter.SimulationFidelityandAudioListener.SimulationFidelity: Turns acoustic simulation on/off on a per-emitter or per-listener basis- By default, this is Enabled.
-
BasePart.AudioCanCollide: Determines whether this part blocks/reflects audio- By default, parts have audio collisions enabled (except for those that make up the default character model)
-
PhysicalProperties.AudioAbsorption: Determines how strongly a material reflects sound- High absorption means less reflection, which reduces echoes.
-
SoundService.CharacterSoundsUseNewApi: Replaces all of the default characterSounds withAudioPlayers andAudioEmitters, so that they can leverage acoustic simulation.
Acoustic Simulation is also affected by:
-
PhysicalProperties.Density: Also determines how strongly a material occludes sound- Parts which are thicker and have higher density will result in more sound muffling
For more information on using the Roblox Audio API, you can also reference this API overview and our tutorials on adding 3D Audio and Voice Chat.
Known Issues
Click here to read more
The simulation is not perfect, and you may observe that:
- When AudioEmitters diffract (bend around corners), only the single, shortest path will be simulated. This can result in scenarios where the apparent direction of a sound changes dramatically with small amounts of movement.
- Reverb/Reflections are only computed around AudioListeners, so in some situations, AudioEmitters positioned in a different location can sound more or less reverberated than expected.
- For example, if an AudioEmitter is inside a stone cave but an AudioListener is outside, the sound will be unmodified even though the cave should cause there to be more reverb.
- Solid walls currently filter sound a little bit less than we would like.
- Sometimes AudioEmitters are more reverberant than expected for a few moments after creation.
These and any other specific sound characteristics of acoustic simulation are subject to change as we improve on the feature through beta.
We are still tuning the algorithms for performance, especially for lower-end devices. Please note that even after public release, acoustic simulation may work differently (or be completely disabled, in some cases) on some lower-end devices in order to maintain performance and stability.
If you encounter anything else or otherwise have any feedback for us, we’d love to hear about it!
A big thanks to @cognitivetest_306, @DinoCoderMom, @Doctor_Sonar, @PeacefulSquirrel, @ReallyLongArms, @phri, and @therealmotorbikematt for making this possible.
