Vic's Acoustic IR Generator

×

Source Audio Guide

The algorithm compares the Pickup signal to this Mic signal and generates an Impulse Response that makes the Pickup sound like the Mic. Record both signals at the same time into the DAW of your choice (Garageband, Audacity, etc.) or a multi-track recorder like a Zoom H5. Ensure both files are max volume without peaking, time-aligned, and from the same performance for best results. Output to mono wav files. 45-60 seconds is plenty of audio for this process.

Pickup WAV

This is the sterile, direct audio signal recorded straight from your acoustic guitar's built-in (usually piezo) pickup. This is the baseline sound that the algorithm will learn to "convert".

Mic WAV

This is the target "studio" sound recorded with a real microphone placed in front of your acoustic guitar. A nice dynamic mic like a Shure SM57 is recommended for best results. In my experience, condensers can be too boomy in the mid range and prone to feedback.

×

Configuration Guide

Algorithm

Standard (Original): The original algorithm that generates a single, polished IR. It uses a chunking method to reject outliers and applies a final Graphic EQ matching process.

Cuki's Algorithm: An alternative algorithm based on Cuki's open-source project. It returns a zip file containing multiple IR variations (Standard, M-file, Blended) so you can choose the one that works best for your instrument. The "M" is Cuki's modified approach to the original algorithm and is often preferred for acoustic guitar IRs.

IR Length (Samples)

Determines the length of the generated impulse response. 2048 represents roughly 46 milliseconds of audio and is the sweet spot for avoiding unnatural low-end ringing caused by mathematically forcing the IR to match the microphone's Graphic EQ. Increasing length (e.g. 4096) allows for finer detail but can introduce muddy bass artifacts.

Smoothing (Fractional Octave)

Smoothes the microscopic mathematical peaks and valleys in the frequency response before generating the final audio. A setting of 1/3 mathematically averages adjacent frequencies, preventing a spiky "comb filter" effect in the high end, resulting in a more natural, wooden acoustic tone. Lowering this value will leave more of those sharp, resonant microscopic peaks intact (potentially introducing harsh highs), while raising this value will more aggressively average the frequencies, potentially dulling the unique character and definition of the acoustic tone.