References of "Embrechts, Jean-Jacques"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailEnhancing Cover Song Identification with Hierarchical Rank Aggregation
Osmalsky, Julien ULg; Van Droogenbroeck, Marc ULg; Embrechts, Jean-Jacques ULg

in Proceedings of the 17th International for Music Information Retrieval Conference (2016, August)

Abstract Cover song identification involves calculating pairwise similarities between a query audio track and a database of reference tracks. While most authors make exclusively use of chroma features ... [more ▼]

Abstract Cover song identification involves calculating pairwise similarities between a query audio track and a database of reference tracks. While most authors make exclusively use of chroma features, recent work tends to demonstrate that combining similarity estimators based on multiple audio features increases the performance. We improve this approach by using a hierarchical rank aggregation method for combining estimators based on different features. More precisely, we first aggregate estimators based on global features such as the tempo, the duration, the loudness, the beats, and the average chroma vectors. Then, we aggregate the resulting composite estimator with four popular state-of-the-art methods based on chromas as well as timbre sequences. We further introduce a refinement step for the rank aggregation called “local Kemenization” and quantify its benefit for cover song identification. The performance of our method is evaluated on the Second Hand Song dataset. Our experiments show an significant improvement of the performance, up to an increase of more than 200 % of the number of queries identified in the Top-1, compared to previous results. [less ▲]

Detailed reference viewed: 28 (5 ULg)
Full Text
See detailReview on the applications of directional impulse responses in room acoustics
Embrechts, Jean-Jacques ULg

in Actes du CFA 2016 (2016, April)

Detailed reference viewed: 26 (0 ULg)
Full Text
See detailEffects of acoustic degradations on cover song identification systems
Osmalsky, Julien ULg; Embrechts, Jean-Jacques ULg

in International Congress on Acoustics: ICA 2016 (2016)

Cover song identification systems deal with the problem of identifying different versions of an audio query in a reference database. Such systems involve the computation of pairwise similarity scores ... [more ▼]

Cover song identification systems deal with the problem of identifying different versions of an audio query in a reference database. Such systems involve the computation of pairwise similarity scores between a query and all the tracks of a database. The usual way of evaluating such systems is to use a set of audio queries, extract features from them, and compare them to other tracks in the database to report diverse statistics. Databases in such research are usually designed in a controlled environment, with relatively clean audio signals. However, in real life conditions, audio signals can be seriously modified due to acoustic degradations. For example, depending on the context, audio can be modified by room reverberation, or by added hands clapping noise in a live concert, etc. In this paper, we study how environmental audio degradations affect the performance of several state-of-the-arty cover song recognition systems. In particular, we study how reverberation, ambient noise and distortion affect the performance of the systems. We further investigate the effect of recording or playing music through a smartphone for music recognition. To achieve this, we use an audio degradation toolbox to degrade the set of queries to be evaluated. We propose a comparison of the performance achieved with cover song identification systems based on several harmonic and timbre features under ideal and noisy conditions. We demonstrate that the performance depends strongly on the degradation method applied to the source, and quantify the performance using multiple statistics. [less ▲]

Detailed reference viewed: 16 (5 ULg)
Full Text
Peer Reviewed
See detailCombining Features for Cover Song Identification
Osmalsky, Julien ULg; Embrechts, Jean-Jacques ULg; Foster, Peter et al

in 16th International Society for Music Information Retrieval Conference (2015, October)

In this paper, we evaluate a set of methods for combining features for cover song identification. We first create multiple classifiers based on global tempo, duration, loudness, beats and chroma average ... [more ▼]

In this paper, we evaluate a set of methods for combining features for cover song identification. We first create multiple classifiers based on global tempo, duration, loudness, beats and chroma average features, training a random forest for each feature. Subsequently, we evaluate standard combination rules for merging these single classifiers into a composite classifier based on global features. We further obtain two higher level classifiers based on chroma features: one based on comparing histograms of quantized chroma features, and a second one based on computing cross-correlations between sequences of chroma features, to account for temporal information. For combining the latter chroma-based classifiers with the composite classifier based on global features, we use standard rank aggregation methods adapted from the information retrieval literature. We evaluate performance with the Second Hand Song dataset, where we quantify performance using multiple statistics. We observe that each combination rule outperforms single methods in terms of the total number of identified queries. Experiments with rank aggregation me- thods show an increase of up to 23.5 % of the number of identified queries, compared to single classifiers. [less ▲]

Detailed reference viewed: 93 (24 ULg)
Full Text
See detailDescription of the research project CIMEDE for the industrial construction of evolutionary, sustainable and economic houses
Duthoit, Fabienne ULg; Embrechts, Jean-Jacques ULg

in Proceedings of the Euronoise 2015 Congress (2015, June)

CIMEDE is a research project supported by the Region of Wallonia (Belgium). It stands for CIMEDE “Construction Industrielle de Maisons Evolutives, Durables et Economiques – industrial construction of ... [more ▼]

CIMEDE is a research project supported by the Region of Wallonia (Belgium). It stands for CIMEDE “Construction Industrielle de Maisons Evolutives, Durables et Economiques – industrial construction of evolutionary, sustainable and economic houses”. Since the materials envisaged for this project are timber and gypsum fibre board, it was necessary to design a constructive principle of lightweight timber frame constructions which meets the industrial, sustainable, evolutionary and economic requirements and also the requirements of the Belgian acoustic standard NBN S01-400-1. First of all, an inventory of constructive elements that are currently used in this type of construction (floors, walls and facades) is presented with their acoustic performances. Then, the laboratory measurements of some samples of floors, facades, interior walls and partition walls are described and discussed. On the basis of the measurement results, several solutions that best meet all criteria were retained and a mock-up was built with these constructive elements. Measurement results on this mock-up are also presented. The final solution for the floor, the partition wall, the interior wall and the facade which meets the acoustic requirements is finally described [less ▲]

Detailed reference viewed: 48 (3 ULg)
Full Text
See detailMeasurement of 3D Room Impulse Responses with a Spherical Microphone Array
Embrechts, Jean-Jacques ULg

in Proceedings of the Euronoise 2015 Congress (2015, June)

Directional room impulse responses (DRIRs) are composed of the sound contributions reaching a given location in the room from a well-defined direction in space. DRIRs can be useful in many applications ... [more ▼]

Directional room impulse responses (DRIRs) are composed of the sound contributions reaching a given location in the room from a well-defined direction in space. DRIRs can be useful in many applications, such as the evaluation of spatial room acoustics parameters, the detection of unwanted specular reflections or the 3D auralization of acoustic spaces. A spherical array containing 16 microphones has been realized to measure DRIRs. The logarithmic sinesweep technique is first applied to measure 16 impulse responses, one for each microphone. A spherical harmonics (SH) decomposition of the sound field is then obtained. Spatial aliasing, placement errors and the ‘white noise gain’ (WNG) have been analysed to define the useful bandwith of this measure, i.e. [250Hz – 4kHz]. The coefficients of the SH decomposition are then processed by some beamforming methods, in order to compute the DRIR in any direction around the spherical array. Time and 3D space representations can be generated. The results obtained in some rooms are illustrated in this paper: it is shown that the combination of the ‘delay-and-sum’ and ‘minimum-variance distorsionless response’ beamforming methods is particularly well suited for the analysis of DRIRs. [less ▲]

Detailed reference viewed: 108 (1 ULg)
Full Text
Peer Reviewed
See detailA Geometrical Acoustics Approach Linking Surface Scattering and Reverberation in Room Acoustics
Embrechts, Jean-Jacques ULg

in Acta Acustica United with Acustica (2014), 100(5), 864-879

A general model of the influence of surface scattering on the reverberation time would have several applications in room acoustics. Such a model is not yet available, and it is the purpose of this paper ... [more ▼]

A general model of the influence of surface scattering on the reverberation time would have several applications in room acoustics. Such a model is not yet available, and it is the purpose of this paper to investigate a geometrical acoustics approach. Starting from the radiative transfer equation, an exponential solution is first developed for the reverberation energy decay in rooms with diffusely reflecting boundaries. Differences with the diffuse sound field (Sabine) theory are highlighted, leading to a modified formula for the reverberation time which is shown to be more in accordance with ray tracing simulations. A general model is then proposed for rooms in which specular and diffuse reflections coexist. This general model is applied in rooms where the specular contribution can be assumed to be quasi-isotropic and uniform. Under this assumption, the reverberation decay is represented by the sum of two exponential functions, depending on the scattering coefficients. However, it is shown that the influence of surfaces’ scattering on reverberation is rather limited in this case. On the contrary, rooms with a pair of parallel surfaces are prone to create significant anisotropy in the cloud of image sources. An analytical formulation is proposed in this case for the specular and diffuse contributions, provided that some assumptions are again made on the specularly reflected sound field. The final expression is not really intuitive concerning the relation between scattering coefficients and reverberation, but it contains all the variables influencing this relation. This allows fast evaluations of the effect of surface scattering in particular situations. Finally, the application of this model to room acoustics computer simulations is illustrated by an example. [less ▲]

Detailed reference viewed: 67 (4 ULg)
Full Text
See detailDesign of a spherical array of microphones for room acoustics applications.
Feron, Hermine; Embrechts, Jean-Jacques ULg

Conference (2014, May 28)

Detailed reference viewed: 38 (4 ULg)
Full Text
Peer Reviewed
See detailPerformances of low-level audio classifiers for large-scale music similarity
Osmalsky, Julien ULg; Van Droogenbroeck, Marc ULg; Embrechts, Jean-Jacques ULg

in International Conference on Systems, Signals and Image Processing (2014, May)

This paper proposes a survey of the performances of binary classifiers based on low-level audio features, for music similarity in large-scale databases. Various low-level descriptors are used individually ... [more ▼]

This paper proposes a survey of the performances of binary classifiers based on low-level audio features, for music similarity in large-scale databases. Various low-level descriptors are used individually and then combined using several fusion schemes in a content-based audio retrieval system. We show the performances of the classifiers in terms of pruning and loss and we demonstrate that some combination schemes achieve a better performance at a minimum computational cost. [less ▲]

Detailed reference viewed: 92 (35 ULg)
Full Text
Peer Reviewed
See detailImprovement in non-linear guitar loudspeaker sound reproduction
Schmitz, Thomas ULg; Embrechts, Jean-Jacques ULg

in International Conference on Systems, Signals and Image Processing (2014, May)

This paper proposes a study of the accuracy of a guitar amplifier loudspeaker simulation. The simulation is based on a non-linear convolution of a signal using Volterra kernels, which are measured in ... [more ▼]

This paper proposes a study of the accuracy of a guitar amplifier loudspeaker simulation. The simulation is based on a non-linear convolution of a signal using Volterra kernels, which are measured in anechoic conditions with a sine-sweep technique. In this paper, we propose an evaluation of the method to minimise the cost in CPU load, while keeping the best performance in the sound reproduction. To assess the performance of the method, we measure errors between the simulated and real sounds. Human listening tests are moreover proposed in order to determine the minimum level of accuracy leading to unaudible differences with the real loudspeaker.. [less ▲]

Detailed reference viewed: 80 (16 ULg)
Full Text
Peer Reviewed
See detailSubjective Evaluation of the Audiovisual Spatial Congruence in the Case of Stereoscopic-3D Video and Wave Field Synthesis
André, Cédric ULg; Corteel, Etienne; Embrechts, Jean-Jacques ULg et al

in International Journal of Human-Computer Studies (2014), 72(1), 23-32

While 3D cinema is becoming increasingly established, little effort has focused on the general problem of producing a 3D sound scene spatially coherent with the visual content of a stereoscopic-3D (s-3D ... [more ▼]

While 3D cinema is becoming increasingly established, little effort has focused on the general problem of producing a 3D sound scene spatially coherent with the visual content of a stereoscopic-3D (s-3D) movie. The perceptual relevance of such spatial audiovisual coherence is of significant interest. In this paper, a subjective experiment is carried out where an angular error between an s-3D video and a spatially accurate sound reproduced through Wave Field Synthesis (WFS) is simulated. The psychometric curve is measured with the method of constant stimuli, and the threshold for bimodal integration is estimated. The impact of the presence of background noise is also investigated. A comparison is made between the case without any background noise and the case with an SNR of 4 dB(A). Estimates of the thresholds and the slopes, as well as their confidence intervals, are obtained for each level of background noise. When background noise was present, the point of subjective equality (PSE) was higher (19.4° instead of 18.3°) and the slope was steeper (-0.077 instead of -0.062 per degree). Because of the overlap between the confidence intervals, however, it was not possible to statistically differentiate between the two levels of noise. The implications for the sound reproduction in a cinema theater are discussed. [less ▲]

Detailed reference viewed: 131 (22 ULg)
Full Text
See detailCapture of 3D sound with a spherical microphone array
Feron, Hermine; Embrechts, Jean-Jacques ULg

Conference (2013, December 05)

Detailed reference viewed: 38 (5 ULg)
Full Text
Peer Reviewed
See detailA New Validated Method for Improving the Audiovisual Spatial Congruence in the Case of Stereoscopic-3D Video and Wave Field Synthesis
André, Cédric ULg; Corteel, Etienne; Embrechts, Jean-Jacques ULg et al

in 3D Imaging (IC3D), 2013 International Conference on (2013, December)

While 3D cinema is becoming increasingly established, little effort has focused on the general problem of producing a 3D sound scene spatially coherent with the visual content of a stereoscopic-3D (s-3D ... [more ▼]

While 3D cinema is becoming increasingly established, little effort has focused on the general problem of producing a 3D sound scene spatially coherent with the visual content of a stereoscopic-3D (s-3D) movie. The perceptual relevance of such spatial audiovisual coherence is of significant interest. In this paper, we explain why the combination of accurate sound positioning and stereoscopic-3D images can lead to an incongruence between the sound and the image for multiple spectators. Then, we adapt to s-3D viewing a method originally proposed for 2D images in the literature to reduce this error. Finally, a subjective experiment is carried out to prove the efficiency of the method. [less ▲]

Detailed reference viewed: 49 (3 ULg)
Full Text
See detailThe contributions of pairs of parallel surfaces in a simple analytical model of room reverberation
Embrechts, Jean-Jacques ULg

in I.C.A. (Ed.) Proceedings of the 2013 ICA Congress (2013, June)

Detailed reference viewed: 44 (15 ULg)
Full Text
See detailPrediction and measurement of the random incidence scattering coefficient of periodic reflective rectangular diffuser profiles
Schmich-Yamane, Isabelle; Embrechts, Jean-Jacques ULg; Müller-Trapet, Markus et al

in Proceedings of the 2013 ICA Congress (2013, June)

Detailed reference viewed: 62 (1 ULg)
Full Text
See detailNonlinear guitar loudspeaker simulation
Schmitz, Thomas ULg; Embrechts, Jean-Jacques ULg

in Proceedings of the AES 134 th convention (2013, May)

Detailed reference viewed: 125 (24 ULg)
Full Text
Peer Reviewed
See detailEfficient database pruning for large-scale cover song recognition
Osmalsky, Julien ULg; Pierard, Sébastien ULg; Van Droogenbroeck, Marc ULg et al

in International Conference on Acoustics, Speech, and Signal Processing (ICASSP) (2013, May)

This paper focuses on cover song recognition over a large dataset, potentially containing millions of songs. At this time, the problem of cover song recognition is still challenging and only few methods ... [more ▼]

This paper focuses on cover song recognition over a large dataset, potentially containing millions of songs. At this time, the problem of cover song recognition is still challenging and only few methods have been proposed on large scale databases. We present an efficient method for quickly extracting a small subset from a large database in which a correspondence to an audio query should be found. We make use of fast rejectors based on independent audio features. Our method mixes independent rejectors together to build composite ones. We evaluate our system with the Million Song Dataset and we present composite rejectors offering a good trade-off between the percentage of pruning and the percentage of loss. [less ▲]

Detailed reference viewed: 164 (56 ULg)
Full Text
Peer Reviewed
See detailA Diffraction Model for Acoustical Ray-Tracing Based on the Energy Flow Lines Concept
Billon, Alexis; Embrechts, Jean-Jacques ULg

in Acta Acustica United with Acustica (2013), 99

Geometrical acoustics models are currently the most popular prediction tools in room-acoustics due to their low computing load. However, they seldom take into account the diffraction occurring at free ... [more ▼]

Geometrical acoustics models are currently the most popular prediction tools in room-acoustics due to their low computing load. However, they seldom take into account the diffraction occurring at free edges. Moreover, the existing diffraction models implemented in geometrical acoustics algorithms are either limited to specularly reflected paths or either present excessive computation load. Recently, a diffraction model based on an approximation of the far-field direction of the Poynting’s vector around half-planes has been introduced for computing the scattering due to faceted dielectric objects. In this article, this model is developed to handle obliquely incident waves and is implemented within an acoustic ray-tracing software using the analogy between the Poynting’s and the acoustic intensity vectors. The proposed model can handle both diffuse and specular reflections and sets no limit in terms of reflection or diffraction order. The first results obtained by ray-tracing agree well with the uniform theory of diffraction and the boundary elements method for single and double diffraction problems even if the interference effects are neglected and as long as the number of emitted rays is sufficient. Moreover, the additional computation load to handle diffraction is shown to be low compared to the one from a classical ray-tracing algorithm. [less ▲]

Detailed reference viewed: 48 (1 ULg)