ABSTRACT
Music signals are highly structured in terms of harmony and rhythm. Content-based Music Information Retrieval (MIR) deals with extracting and processing meaningful information from music audio signals. We first present a model for the estimation of musical content. We propose an innovative approach for music description at several time-scales in a single unified formalism. More specifically, chord information at the analysis-frame level and global semantic structure are integrated in an elegant and flexible model. Using Markov Logic Networks (MLNs), low-level signal features are encoded with high-level information expressed by logical rules, without the need of a transcription step. Our results demonstrate the potential of MLNs for music analysis, as they can express both structured relational knowledge through logic as well as uncertainty through probabilities. We then investigate how this extracted music content information can be used in the context of structured sparse expansion of audio signals of music on hybrid dictionaries. Through the denoising task application, we show that musical structure constraints can be used to build models capable of giving a relevant and legible representation of Western tonal music audio signals.