MicAugment: One-shot Microphone Style Transfer

TL;DR

MicAugment is a novel one-shot microphone style transfer method that enhances audio model robustness by synthesizing audio under target device conditions using minimal target audio data.

Contribution

It introduces a new approach for one-shot microphone style transfer that identifies and applies device-specific transformations to improve robustness.

Findings

Successfully applies style transfer to real audio

Significantly improves downstream model robustness

Effective as a data augmentation technique

Abstract

A crucial aspect for the successful deployment of audio-based models "in-the-wild" is the robustness to the transformations introduced by heterogeneous acquisition conditions. In this work, we propose a method to perform one-shot microphone style transfer. Given only a few seconds of audio recorded by a target device, MicAugment identifies the transformations associated to the input acquisition pipeline and uses the learned transformations to synthesize audio as if it were recorded under the same conditions as the target audio. We show that our method can successfully apply the style transfer to real audio and that it significantly increases model robustness when used as data augmentation in the downstream tasks.