SubSpectral Normalization for Neural Audio Data Processing

TL;DR

This paper introduces SubSpectral Normalization (SSN), a novel normalization technique that splits the frequency domain of audio data into sub-bands and normalizes each separately, enhancing neural network performance on audio tasks.

Contribution

The paper proposes SSN, a new normalization method that handles frequency sub-bands differently, addressing the unique characteristics of audio spectrograms in neural networks.

Findings

SSN improves neural network performance on audio data.

SSN effectively removes inter-frequency deflection.

SSN enhances frequency-aware feature learning.

Abstract

Convolutional Neural Networks are widely used in various machine learning domains. In image processing, the features can be obtained by applying 2D convolution to all spatial dimensions of the input. However, in the audio case, frequency domain input like Mel-Spectrogram has different and unique characteristics in the frequency dimension. Thus, there is a need for a method that allows the 2D convolution layer to handle the frequency dimension differently. In this work, we introduce SubSpectral Normalization (SSN), which splits the input frequency dimension into several groups (sub-bands) and performs a different normalization for each group. SSN also includes an affine transformation that can be applied to each group. Our method removes the inter-frequency deflection while the network learns a frequency-aware characteristic. In the experiments with audio data, we observed that SSN can efficiently improve the network's performance.