It predicts frequency coefficients that allow the dynamic range compression, via a trained deep neural network, and applies them to unmastered audio signal served as input. In this work we present a system capable of modelling such behavior focusing on the automatic dynamic range compression. Modeling such dynamic operations, which involve adaptation regarding the audio content, becomes vital in automated applications since it significantly affects the overall performance. This operation relies on musical and perceptually pleasing facets of the perceived acoustic characteristics, transmitted from the audio material under the mastering process. With respect to the genre and style of the audio content, the parameters of these techniques are controlled by a mastering engineer, in order to process the original audio material. The process of audio mastering often, if not always, includes various audio signal processing techniques such as frequency equalization and dynamic range compression. Improvements were also observed when compared to relevant and commercial software.
Results from conducted listening tests, incorporating professional music producers and audio mastering engineers, demonstrate on average an equivalent performance compared to professionally mastered audio content. Both dynamic range compression and the prediction of the corresponding frequency coefficients take place inside the time-frequency domain, using magnitude spectra acquired from a critical band filter bank, similar to humans’ peripheral auditory system.
JO - Journal of the Audio Engineering SocietyĪB - The process of audio mastering often, if not always, includes various audio signal processing techniques such as frequency equalization and dynamic range compression. TI - Deep Neural Networks for Dynamic Range Compression in Mastering Applications
Improvements were also observed when compared to relevant and commercial stylianos ioannis and drossos, konstantinos and virtanen, tuomas and schuller, gerald},
doi:Ībstract: The process of audio mastering often, if not always, includes various audio signal processing techniques such as frequency equalization and dynamic range compression. Schuller, "Deep Neural Networks for Dynamic Range Compression in Mastering Applications," Paper 9539, (2016 May.). Music Production for Emerging Audio Formats.Audio for Virtual and Augmented Reality.