Timbre Analysis of Music Audio Signals with Convolutional Neural Networks
This work addresses the challenge of efficiently modeling timbre in music audio for applications in music information retrieval, but it is incremental as it builds on existing CNN trends.
The authors tackled the problem of learning timbre representations from audio signals by proposing a design strategy for Convolutional Neural Networks (CNNs) to capture relevant time-frequency contexts, which was successfully assessed for tasks like singing voice phoneme classification, musical instrument recognition, and music auto-tagging.
The focus of this work is to study how to efficiently tailor Convolutional Neural Networks (CNNs) towards learning timbre representations from log-mel magnitude spectrograms. We first review the trends when designing CNN architectures. Through this literature overview we discuss which are the crucial points to consider for efficiently learning timbre representations using CNNs. From this discussion we propose a design strategy meant to capture the relevant time-frequency contexts for learning timbre, which permits using domain knowledge for designing architectures. In addition, one of our main goals is to design efficient CNN architectures -- what reduces the risk of these models to over-fit, since CNNs' number of parameters is minimized. Several architectures based on the design principles we propose are successfully assessed for different research tasks related to timbre: singing voice phoneme classification, musical instrument recognition and music auto-tagging.