Auto-adaptive Resonance Equalization using Dilated Residual Networks
This work addresses the need for automated tools to assist sound engineers in music and audio production, though it is incremental as it builds on existing deep learning methods for audio processing.
The authors tackled the problem of automating resonance equalization in audio production by developing a two-component system: a dynamic equalizer for detecting resonances and a deep neural network to predict optimal attenuation factors based on audio input. They found that a dilated residual network performed as well as a feature-based architecture, both significantly outperforming a baseline in predicting human-preferred attenuation factors.
In music and audio production, attenuation of spectral resonances is an important step towards a technically correct result. In this paper we present a two-component system to automate the task of resonance equalization. The first component is a dynamic equalizer that automatically detects resonances and offers to attenuate them by a user-specified factor. The second component is a deep neural network that predicts the optimal attenuation factor based on the windowed audio. The network is trained and validated on empirical data gathered from an experiment in which sound engineers choose their preferred attenuation factors for a set of tracks. We test two distinct network architectures for the predictive model and find that a dilated residual network operating directly on the audio signal is on a par with a network architecture that requires a prior audio feature extraction stage. Both architectures predict human-preferred resonance attenuation factors significantly better than a baseline approach.