Skip to guide
← Tool guides

Audio, explained · 16 / 36

FIR Filter

Apply the response of an imported impulse file.

Visual walkthrough3 configuration controlsSynced audio walkthrough
See how it works ↓

Choose WAV or MP3. The backend decodes at the graph rate, uses the first two seconds and normalizes to unit energy. Processing declares 512 samples of latency; wet and dry paths are equally delayed. Without a response file, audio passes unchanged.

01 / Understand

The idea, before the settings.

Choose an impulse response WAV or MP3. AudioRouter uses up to the first two seconds, normalizes it to unit energy, and convolves it with the input. Wet mix blends the filtered result with an equally delayed dry path; Output gain sets the final level. Without a selected file, the node passes audio unchanged.

SourceFIR FilterMeter → Output

Use it in AudioRouter

  1. Choose and approve an impulse-response WAV in the actual FIR inspector.
  2. Wait for the prepared response and inspect its duration, partitions and fixed latency.
  3. Begin with Wet 100% and Gain 0 dB. Compare against dry at matched loudness, then blend if useful.
  4. If loading fails, inspect the file/error before saving the intended route.

What success looks like

The source is convolved with the loaded response. Wet/Dry blends aligned paths; output Gain adjusts the result.

If it does not work

A long or strongly colored response can smear attacks or increase level. Check the IR and headroom before blaming the source.

Select the node to open Properties. Click its name to rename it. While playing, supported parameter-only edits save after a short pause. For structural edits, use Save when the canvas marks a change. Enabled and Bypass are distinct live controls; see the in-app help for their behavior.

02 / Explore

Make the behavior visible.

Browser convolution uses a short built-in multi-tap response, not your file. Compare Dry with Wet to hear the tail; the actual inspector's load/prepare/partition workflow is described below.

The example repeats until you press Stop, and control changes are heard immediately. The picture shows the tool's shape rather than time; follow the timing in Follow the sound below. This is not a DSP measurement or control for your running AudioRouter session.

03 / Configure

Every control, with a reason.

01

Choose WAV or MP3

mediaId

Choose the impulse response through the file picker. The backend imports media and the editor stores its media identity; fileName is the display name, not a second effect control.

Range / choicesDepends on your selectionDefaultChoose explicitly

↓ Lower / alternativeNo numeric increase/decrease applies to file selection.

↑ Higher / alternativeChoose a different response to change the filtering behavior.

Try thisChoose a short WAV impulse response, then compare 100% wet with 50% wet.

Keep in mind Only the first two seconds are used and normalized to unit energy. Without a chosen response the tool passes unchanged.

02

Wet mix

wetPercent

Blend between filtered and equally delayed dry sound.

Range / choices0 % → 100 %Default100 %

↓ Lower / alternativeIncludes more dry sound.

↑ Higher / alternativeIncludes more convolution output.

Try this100% is only filtered sound; 50% blends the two.

Keep in mind Without an impulse file the node passes unchanged.

03

Output gain

gainDb

A level adjustment expressed in decibels.

Range / choices-24 dB → 12 dBDefault0 dB

↓ Lower / alternativeMore negative values make it quieter.

↑ Higher / alternativeMore positive values make this output louder.

Try thisTry −6 dB to create headroom.

Keep in mind For FIR this is output gain after filtering; an upstream overload is not repaired.

04 / Put it to work

Start with a useful example.

Example 1

Explore a supplied response

  1. Choose a WAV impulse response you have permission to use.
  2. Start Wet mix at 100% and Output gain at 0 dB.
  3. Compare bypass, then lower Wet mix for a blend.
Example 2

Match loudness

  1. With a selected response, compare the same passage through and around the tool.
  2. Use Output gain to reduce an unexpectedly loud result.
  3. Watch downstream peaks as well as perceived loudness.

These are starting points. Compare at a comfortable monitoring level and adjust for your own source.

05 / Get unstuck

Small details that matter.

  • A song file is not usually a useful impulse response.
  • Only the first two seconds are used.
  • Unit-energy normalization does not guarantee matching perceived loudness.