Why do the spaces not have long tails?

Because the cue that tells a listener how big a room is comes from the first reflections, not from the decay. Keeping the tail under a quarter of a second in every space means dialogue stays intelligible while still sounding as though it happened somewhere, which a full reverb would not allow.

How were the reflection patterns chosen?

They were laid out by hand for each space, with tap times and gains that match the physical description: a hallway has close walls, a living room has soft surfaces that absorb, a stairwell has hard concrete, and a warehouse has far walls with a long distance between them.

Why is the output the same length as the input?

So that the short tail cannot run past the end of the file and leave silence after it. The reflections in these spaces are all under a tenth of a second in most cases, so the truncation removes nothing audible while keeping the file's length predictable.

Is the effect mono or stereo?

Each input channel is convolved with the same impulse response, so a stereo file stays stereo and a mono file stays mono. There is no attempt to build a stereo pair from a mono source, which is why the reflections land in the same place on both sides.

How much reflection should I use?

The default of nine decibels down is a good starting point for dialogue, since the reflections should be felt rather than noticed. Raising the level makes the placement obvious and, on a close-miked voice, can start to sound like a slap rather than like a room.

Hallway / Room Simulator

You Might Also Need

Convolution Reverb

Realistic reverb using impulse responses

Reverb Simulator

Add realistic room reverberation

Stereo Widener

Enhance stereo field width

Room Tone Generator

Generate matching room tone

Hallway Room Simulator adds the first echoes a room makes and nothing after them. Discrete reflections are the cue the ear uses to judge how big a space is, so the sound gains a location without gaining a tail.

What is Hallway / Room Simulator?

Choose one of four spaces, set the reflection level, drop a file, and place it; the early reflections are convolved into the recording and mixed with the dry signal.

The four spaces and the control:

Hallway: taps at nine, thirteen, nineteen and twenty six milliseconds with a short eighty millisecond tail, described as narrow hard walls and dense fast taps.
Living room: taps at twelve, twenty one and thirty one milliseconds with a hundred and twenty millisecond tail, soft and sparse because furniture absorbs.
Stairwell: taps at eight, fourteen, twenty two, thirty three and forty seven milliseconds with a hundred and eighty millisecond tail, loud and ringing because concrete reflects.
Warehouse: taps at twenty five, forty eight, seventy four and a hundred and three milliseconds with a two hundred and fifty millisecond tail, big, bright and sparse.
A reflection level from thirty decibels down to nothing, defaulting to nine decibels down, which is the wet gain.

The taps are laid out by hand for each space, which is why the four sound different rather than like four settings of the same thing. The gaps between the first reflections are what the ear translates into room size: a hallway's nine millisecond first tap says the walls are close, while a warehouse's twenty five millisecond first tap and its hundred and three millisecond last one say the far surface is metres away. The gains fall across the pattern too, since each successive reflection has travelled further and lost more energy. The tail that follows the taps is short by design, between eighty and two hundred and fifty milliseconds, and it exists only to give the reflections something to sit in rather than to be heard as reverberation. That is what makes this tool different from its convolution sibling: a full reverb rings for seconds and smears detail, whereas the reflections here end within a tenth of a second in most cases, so dialogue stays crisp and simply sounds as though it happened somewhere. The mixing is additive, with the wet signal scaled by the reflection level and added to the unchanged dry signal, and the output is deliberately the same length as the input so that the reverb tail cannot run past the end of the file.

What it is not:

No long reverb tail, so it cannot imitate a large space sustaining a sound.
The four spaces have fixed reflection patterns, with no control of individual taps.
No pre-delay, damping or diffusion controls.
The reflection level only scales the whole pattern rather than reshaping it.
Nothing models where a source sits inside the space.
Output is sixteen bit WAV.

Use it for dialogue and voice-over that should sound placed in a room, which is the job early reflections do best, and for any mix that needs a sense of location without the wetness a reverb adds. When a longer, more obviously reverberant space is wanted, the convolution reverb supplies actual room responses, and the reverb simulator gives dialable decay and damping. When the space should instead come from a stereo pair rather than from reflections, the stereo widener works on the image directly.

How to use Hallway / Room Simulator

Drop an audio file onto the panel and let it decode.
Choose one of the four spaces by how large and how hard it should sound.
Set the reflection level, leaving it well below the dry signal for a natural placement.
Place the sound in the room and download the WAV, which is the same length as the source.

When to use Hallway / Room Simulator vs related tools

Hallway Room Simulator is the tool when a recording needs a location rather than a wash of reverb, which is exactly what early reflections provide. Convolution Reverb puts the same recording in a space with a long tail, and Reverb Simulator offers adjustable decay when the tail itself is the point. When the sense of space should come from the stereo image rather than from echoes, Stereo Widener is the closer tool.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.