This question has been bothering me for a while, so the situation is next: imagine that you have a person talking to you surrounded by four walls with an opened door and you are as a listener located outside this four walls. So if that you gotta hear a reverbed signal but you won't hear it as if you were in the same room with the sound source. If there're any popular ways to simulate this kind of situation? Thanks in advance and sorry for my grammar mistakes, i would explain it clearer if needed. enter image description here

  • 1
    Are you doing this for a movie/TV soundtrack? if so you can make a convolution reverb based on even the clapperboard used for the original recording, with something like Altiverb. idk of anything cheaper designed to do the same job, but there probably is.
    – Tetsujin
    May 18, 2019 at 14:38
  • @Tetsujin Thanks for your advice. I would dive into this, hope it would help me. I'm doing it for a game. And this is not required from me yet. But i just want to expand my knowledge.
    – Turbo Ash
    May 18, 2019 at 18:27
  • Have a look at how they do DIY convolution sampling. Altiverb is not cheap... 500 - 800 USD, but I don't know of another system that captures audio spaces so accurately or easily, being able to use the clapperboard from a take inside that space can be a lifesaver for post-pro, if the space is awkward or unusual. - audioease.com/altiverb/sampling.php
    – Tetsujin
    May 18, 2019 at 19:11
  • @Tetsujin Well i checked it. Seems like it's not what I need. I don't want a reverb to match specific elements of a room. Much easier actually. Imagine that you're on the street passing by a garage where a band practice for instance. So the sound in this garage has a lot of reverb. But you're listening to it from outside, so it reflects from the walls of garage, but doesn't reflect from anyting on the street. Like on the picture
    – Turbo Ash
    May 18, 2019 at 19:46

2 Answers 2


Reverb is made up of two major components:

  1. Early Reflections
  2. Late Reflections.

Early reflections are the primary and secondary reflections from surfaces in proximity to the sound source and late reflections are all the remaining random reflections of reflections that occur until all the energy is lost.

As the size and reflectivity of the space changes, so do the parameters that affect both the early and late reflections. Larger spaces such as cathedrals will have more late reflections and a longer decay time. Smaller and less reflective spaces will likely have fewer early and late reflections and a much shorter decay time.

A small, reflective space, will have a high level of early reflections but a small amount of late reflections and a very low overall reverberation/decay time.

When passing a space such as this and hearing what is going on, remember that as you pass, you will hear some level of direct sound, but not a great deal and not for a great amount of time, so bear this in mind when mixing. Also you are most likely to hear mostly early reflections and only a little of the late reflections. Overall reverb time is likely to be very short, which will also affect the amount of late reflections you will experience.

Play around with the parameters in a simple algorithmic reverb plugin, such as Valhalla and also tune the experience for the motion past the opening, by mixing in some of the direct sound as you are directly exposed to the sound source.


Well, if you are using it for a game (as you mention in the comments), maybe you could use the game engine's audio functionality and let it do what it does best..., simulate sound propagation! That is, if you use a game engine (such as Unity, Unreal, or any other game engine that has built-in audio engines).

Whether you use a "default" audio engine or working on one of your own (or someone else's) you could treat the "garage" as a single source, or even a combination of more than one. This could possibly work if the size of your source - the garage that is - is quite big (you are close to it). Adding more sources would definitely be the way to go if you have distinguishable sources you want to simulate inside the garage.

Then, by treating the whole place as a source (this is what it is to an external observer), you could try mixing its direct sound with an artificial reverberation (plug-in, game engine's reverb, custom algorithm, convolution, etc.). Similarly the same approach should be followed if the garage is "made" of many sources.

Mark here has suggested a good way to experiment with the parameters of the reverberation algorithm and what to keep in mind.

In my opinion, it would be best to find a way to automate the process, since you are dealing with a game and many more such situations will most probably occur during gameplay.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.