Automatic Audio Mastering System
Ervk Najduchowski, Marcin Lewandowski, Piotr Bobiński · 2018
The modern music production process requires multiple steps of digital signal processing such as audio frequency response equalization, or dynamic range compression. In order to process the original audio material the mastering engineer controls the parameters of these processing algorithms with respect to genre and style of audio content. The main purpose of this processing is to aesthetically enhance perceived acoustic characteristics of the signals. The selection and the adj ustment of these parameters relies on the continuous interaction between the audio mastering engineer and the apparatus that handles the audio signals. Modelling such dynamic operations becomes very important in automated applications. In this work we present a system which automatically enhances unprocessed audio signal with respect to the specific parameters of the reference audio material. These parameters are obtained through analysis of magnitude spectrum (spectral roll-off point, and energy calculated for specified frequency bands), amplitude histogram, audio content tempo, signal envelope's timing features and LUFS parameter. Results from conducted online listening tests are presented and discussed, along with objective measurements of unprocessed and reference audio signal.