Search

International audience; We consider the problem of blind source separation for underdetermined convolutive mixtures. Based on the multiplicative narrowband approximation in the time-frequency domain with the help of Short-Time-Fourier-Transform (STFT) and the sparse representation of the source signals, we formulate the separation problem into an optimization framework. This framework is then generalized based on the recently investigated convolutive narrowband approximation and the statistics of the room impulse response. Algorithms with convergence proof are then employed to solve the proposed optimization problems. The evaluation of the proposed frameworks and algorithms for synthesized and live recorded mixtures are illustrated. The proposed approaches are also tested for mixtures with input noise. Numerical evaluations show the advantages of the proposed methods.

Pseudointensity vectors (PIVs) provide a means of direction of arrival (DOA) estimation for spherical microphone arrays using only the zeroth and the first-order spherical harmonics. An augmented intensity vector (AIV) is proposed which improves the accuracy of PIVs by exploiting higher order spherical harmonics. We compared DOA estimation using our proposed AIVs against PIVs, steered response power (SRP) and subspace methods where the number of sources, their angular separation, the reverberation time of the room and the sensor noise level are varied. The results show that the proposed approach outperforms the baseline methods and performs at least as accurately as the state-of-the-art method with strong robustness to reverberation, sensor noise, and number of sources. In the single and multiple source scenarios tested, which include realistic levels of reverberation and noise, the proposed method had average error of $1.5^{\circ }$ and $2^{\circ }$ , respectively.

This paper introduces a method to recover unmeasured room impulse responses (RIRs) in acoustical spaces with unknown properties and contents, by means of a compressed sensing methodology. Methods published in the existing literature have been validated in empty, convex rooms; a limited subset of the many, diverse acoustical spaces one can encounter. It results a challenge to represent such diverse wave phenomena with a sparse set of plane waves or equivalent sources, given the coupling between the sparsity of such representations and hypotheses regarding the properties of the acoustical space and its contents, far-field measurement distances, and other parameters. In contrast to this philosophy, the method introduced in this paper exploits the sparsity inherent to the mathematical structure of the wavefronts present in the RIRs, which without further hypotheses carry themselves all the information about the wave propagation in the room. In essence, the measured RIRs are instead represented with a sparse set of curved elementary functions of various sizes, propagation directions and times of arrival, which are linked with the various shapes and locations of the unknown scatterers and boundaries in the room. The main contribution of this work is thus to enable the measurement of RIRs in more complex acoustical spaces, while keeping the number of microphones to a minimum with the use of compressed sensing. The method is formulated as a sparse optimization problem, and the solution is obtained with an iterative thresholding algorithm whose threshold value is determined from the measurements. An analysis of sensing coherence is included, and the performance of the method is experimentally evaluated with 1D microphone array measurements in two lecture rooms and one meet-ing room. For the sake of comparison, the RIRs are also linearly interpolated using a low-pass filter in the wavenumber-frequency domain. The experimental results demonstrate that the proposed method is superior than linear interpolation in all the cases investigated, motivating further development of the method to higher spatial dimensions. In terms of accuracy, the proposed method attains recovery errors in the same order of magnitude as those attained by methods in the literature, yet here the acoustical spaces have arbitrary contents and exhibit more complex geometries and boundary conditions. QC 20190815

International audience; Measuring the Room Impulse Responses within a finite 3D spatial domain can require a very large number of measurements with standard uniform sampling. In this paper, we show that, at low frequencies, this sampling can be done with significantly less measurements, using some modal properties of the room. At a given temporal frequency, a plane wave approximation of the acoustic field leads to a sparse approximation, and therefore a compressed sensing framework can be used for its acquisition. This paper describes three different sparse models that can be constructed, and the corresponding estimation algorithms: two models that exploit the structured sparsity of the soundfield, with projections of the modes onto plane waves sharing the same wavenumber, and one that computes a sparse decomposition on a dictionary of independent plane waves with time / space variable separation. These models are compared numerically and experimentally, with an array of 120 microphones irregularly placed within a 2×2×2 m volume inside a room, with an approximate uniform distribution. One of the most challenging part is the design of estimation algorithms whose computational complexity remains tractable.

Multiple source localization is an important task in acoustic signal processing with applications including dereverberation, source separation, source tracking and environment mapping. When using spherical microphone arrays, it has been previously shown that Pseudo-intensity Vectors (PIV), and Augmented Intensity Vectors (AIV), are an effective approach for direction of arrival estimation of a sound source. In this paper, we evaluate AIV-based localization in acoustic scenarios involving multiple sound sources. Simulations are conducted where the number of sources, their angular separation and the reverberation time of the room are varied. The results indicate that AIV outperforms PIV and Steered Response Power (SRP) with an average accuracy between 5 and 10 degrees for sources with angular separation of 30 degrees or more. AIV also shows better robustness to reverberation time than PIV and SRP.

International audience; Ircam Spat is a software suite dedicated to sound spatialization and artificial reverberation. This paper traces the evolution of the software during the last few years and presents its current state with a focus on the newly available features.

International audience; This paper deals with the interpolation of the Room Impulse Responses (RIRs) within a whole volume, from as few measurements as possible, and without the knowledge of the geometry of the room. We focus on the early reflections of the RIRs, that have the key property of being sparse in the time domain: this can be exploited in a framework of model-based Compressed Sensing. Starting from a set of RIRs randomly sampled in the spatial domain of interest by a 3D microphone array, we propose a modified Matching Pursuit algorithm to estimate the position of a small set of virtual sources. Then, the reconstruction of the RIRs at interpolated positions is performed using a projection onto a basis of monopoles, which correspond to the estimated virtual sources. An extension of the proposed algorithm allows the interpolation of the positions of both source and receiver, using the acquisition of four different source positions. This approach is validated both by numerical examples, and by experimental measurements using a 3D array with up to 120 microphones.

Room acoustics refers to the audio capture of an enclosed room. It can be seen as its color and fully describes the characteristics of the room, including the shape, size or the population. The goal of the project is to set up a database of Room impulse responses which will be useful in many fields in order to test algorithms. This document covers the way to set up equipment and successfully capture one of those room impulse responses.

This thesis is a study of the spatial evolution of the sound field. We first present an analysis of the sound field along different geometries. In the case of the sound field studied along a line in a room, we describe a two-dimensional function characterizing the sound field along space and time. Calculating the Fourier transform of this function leads to a spectrum having a butterfly shape. The spectrum is shown to be almost bandlimited along the spatial frequency dimension, which allows the interpolation of the sound field at any position along the line when a sufficient number of microphones is present. Using this Fourier representation of the sound field, we develop a spatial sampling theorem trading off quality of reconstruction with spatial sampling frequency. The study is generalized for planes of microphones and microphones located in three dimensions. The presented theory is compared to simulations and real measurements of room impulse responses. We describe a similar theory for circular arrays of microphones or loudspeakers. Application of this theory is presented for the study of the angular sampling of head-related transfer functions (HRTFs). As a result, we show that to reconstruct HRTFs at any possible angle in the horizontal plane, an angular spacing of 5 degrees is necessary for HRTFs sampled at 44.1 kHz. Because recording that many HRTFs is not easy, we develop interpolation techniques to achieve acceptable results for databases containing two or four times fewer HRTFs. The technique is based on the decomposition of the HRTFs in their carrier and complex envelopes. With the Fourier representation of the sound field, it is then shown how one can correctly obtain all room impulse responses measured along a trajectory when using a moving loudspeaker or microphone. The presented method permits the reconstruction of the room impulse responses at any position along the trajectory, provided that the speed satisfies a given relation. The maximal speed is shown to be dependent on the maximal frequency emitted and the radius of the circle. This method takes into account the Doppler effect present when one element is moving in the scenario. It is then shown that the measurement of HRTFs in the horizontal plane can be achieved in less than one second. In the last part, we model spatio-temporal channel impulse responses between a fixed source and a moving receiver. The trajectory followed by the moving element is modeled as a continuous autoregressive process. The presented model is simple and versatile. It allows the generation of random trajectories with a controlled smoothness. Application of this study can be found in the modeling of acoustic channels for acoustic echo cancellation or of time-varying multipath electromagnetic channels used in mobile wireless communications.