The Convolver Node performs a linear convolution effect, using an impulse response. In simpler – and practical – terms, it adds an environmental effect, or reverb to an audio signal. That is, it makes the audio signal sound as if it was heard in a reverberation room. Or just any large room in general.
For this, you can either bring your own impulse response (e.g. a field recording), or generate them in the Convolver Node itself. By default, it’s the latter.
Combine the Convolver Node with the Echo Node to create deep, immersive soundscapes, even from simple sounds. This is a common way to add depth to synth pads, or pretty much any sound that can take the additional reverberation.
How to Use
The Convolver Node needs a source audio signal connected to its audio input, to which it can apply the effect, and then output the result. The input audio signal can come from anywhere, as long as it's audible: an Audio File Node, Oscillator Node, Noise Generator Node all work. Click their output, then click the audio input of the Convolver Node to establish a connection.
Connect to an Audio Destination Node to listen to the output, and play your project if your audio source needs it.
An audio signal to which to add reverb. This only supports a mono or stereo signal, otherwise the Convolver Node will down-mix it to stereo.
A control signal to automate the Mix setting, from 0 to 1. You can enable this input from the Node’s properties panel, and is off by default.
Sets how much wet to add to the mix. Greater values emphasize the effect.
Sets how the impulse response is created:
- Generate room impulse response – generates an impulse response by modeling a reverberation room. This mode results in a realistic reverberation feedback, and has plenty of options to customize the result. Ideal to add environmental depth to your sounds.
- Generate white noise – generates an impulse response based on white noise. It results in a soft, immediate reverberation response, which is ideal to quickly add depth to sounds. However, it lacks advanced options to fine-tune the results.
- Open from file – use a file as the impulse response, such as a field recording. Supports stereo (2-channel) and true-matrix (4-channel) impulse responses.
White Noise mode only
Sets the duration of the generated impulse response noise, in seconds. Longer values, of course, result in a longer reverb.
Noise filter cutoff
The generated impulse response noise is processed by a lowpass filter under the hood. This makes the reverberation more realistic, and less noisy.
This setting controls the cutoff frequency of this internal lowpass filter. Higher values result in more perceived noise in the output, depending on the connected audio source.
Noise fade curve
Sets the decay (or fade-out) curve of the generated noise impulse response. A value of 0 results in a completely linear decay. Higher values result in a steeper decay initially, tapering down to a slower decay towards the end. Higher values are generally more “realistic”, while lower values result in more of a “smeared” audio output.
Acts as a seed for the internal random number generator. Different values result in a different impulse response. This setting is unlikely to cause any perceivable change to the resulting audio, especially with a single Convolver Node.
Room Impulse Response mode only
The basic dimensions of the modeled reverberation room, in meters (width, height, and length). A larger room size results in a longer reverb effect, and a sharper initial reverberation feedback. Note, that the actual dimensions also depend on the room shape setting.
How much sound to be reflected from the reverberation room walls, ceiling, and floor. A higher value will result in a longer reverb effect, and a more convoluted audio output. Note, that generating an impulse response, with a very high Surface Reflection value and large room size, can take a long time to complete (up to 10 seconds or even more).
Distance / Delay
How far apart the simulated microphone and simulated audio source are placed. This is a percentage of the room length. Increasing this value will emphasize the reverberation effect, and add a delay before you hear the first reverberation feedback.
Moves the simulated microphone, and the simulated sound source around. This adds a stereo effect, turning even simple mono audio sources to rich stereo sounds.
Controls the reverberation room shape. Lower values produce a box-like room, while higher values produce a longer, hallway-like room instead.
This setting doesn’t affect the main reverberation characteristics – such as the total duration – as much. However, different room shapes provide significantly different results in the audio output.
The audio signal with the convolution/reverb effect. This output is always stereo, except when you use a mono impulse response file, with a mono audio input. Extra channels in the input are automatically mixed to stereo.
Before applying the convolution effect to the connected Audio line, the Convolver Node first normalizes the impulse response file. This uses equal-power normalization.
The White noise mode creates a full-spectrum stereo white noise with a full-length decay (fade-out). In this mode, the generated impulse response is deterministic, meaning the same settings always result in the same result. You can tune this using the Noise variation setting. This setting acts as the seed for the internal random number generator.
Long impulse responses (either generated or file-based) are computationally expensive. Because of this, you should generally use them sparsely. For example, instead of using a Convolver Node multiple times, consider using only 1, after merging your audio using an Add (Audio/Control) Node.