AI vs Sound Engineers in Reverberation
On the Jan 16th, Manni CHEN and Dr PerMagnus LINDBORG published an article entitled “Observations on Guitar Music Produced by AI Reverberation and Professional Sound Engineer ” in the International Journal of Music Science, Technology and Art (IJMSTA) (Vol. 5(1) – January 2023).
This paper observed the comparisons in terms of the reverberation produced by AI engines and professional sound engineers via extracting the audio features and conducting listening tests, after which the statistical methods were applied to analyse the collected data from objective and subjective domains. This work has the potential to be a valuable contribution as it can shed light on how humans perceive AI-generated audio effects such as reverberation differently to human-generated ones.
Welcome to check it out!
http://www.ijmsta.com/archive_issue.php?volume=5&issue=1
Abstract – Artificial intelligence (AI) technologies have been applied in music production to create various sound effects, including reverberation. However, observation on such applications has not yet been fully explored in research studies. This paper reports results from a study comparing reverberation processing on six recordings of guitars, with musical phrasing, made by an AI software and two professional sound engineers. Audio features were extracted using the MIR Toolbox, and perceptual ratings on semantic scales were collected in two listening tests (N = 10, N = 33). Logistic regression was carried out on the two datasets in parallel. An increase in perceived Wetness or decrease in perceived Clarity was associated with a higher probability that the reverberation was made by the AI rather than a Human. For extracted audio features, lower Brightness, Rolloff, and Centroid, which are all indicators for a darker, low frequency emphasized sound, were more likely made by the AI. This study contributes to an understanding of the differences between AI- and human-generated audio effects used in music production.
DOI: https://doi.org/10.48293/IJMSTA-93
