Optimal spectral transportation with application to music transcription

Authors: Rémi Flamary, Cédric Févotte, Nicolas Courty, Valentin Emiya

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental A very fast and simple decomposition algorithm that achieves state-of-the-art performance on real musical data. and 6 Experiments
Researcher Affiliation Academia Rémi Flamary Université Côte d Azur, CNRS, OCA remi.flamary@unice.fr Cédric Févotte CNRS, IRIT, Toulouse cedric.fevotte@irit.fr Nicolas Courty Université de Bretagne Sud, CNRS, IRISA courty@univ-ubs.fr Valentin Emiya Aix-Marseille Université, CNRS, LIF valentin.emiya@lif.univ-mrs.fr
Pseudocode No The paper describes the proposed algorithm steps in prose but does not provide structured pseudocode or an algorithm block.
Open Source Code Yes A Python implementation of OST and real-time demonstrator are available at https://github. com/rflamary/OST
Open Datasets Yes We consider in this section the transcription of a selection of real piano recordings, obtained from the MAPS dataset (Emiya et al., 2010).
Dataset Splits Yes Half of the recording is used for validation of the hyper-parameters and the other half is used as test data.
Hardware Specification No The paper mentions results were run 'on an average desktop PC' but does not provide specific hardware details such as CPU/GPU models or memory.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies.
Experiment Setup Yes For PLCA, we validated 4 and 3 values of the width and amplitude dampening of the Gaussian kernels used to synthesise the dictionary. For OST, we set ϵ = qϵ0 in Eq. (4), which was found to satisfactorily improve the discrimination of octaves increasingly with frequency, and validated 5 orders of magnitude of ϵ0. For OSTe, we additionally validated 4 orders of magnitude of λe.