Number Of Sound Sources Research Articles

Loudness of an auditory scene

When multiple sound sources, each of equal level, are added together to create an auditory scene, the overall level of the scene grows with the increased number of sources despite the fact that the level of the individual sound sources remains the same. This experiment addresses a simple question: how does the overall perceived loudness of an auditory scene change as the number of sound sources in the scene changes. Sixteen Consonant-Vowels (CVs), spoken by ten males and ten females, were the sound stimuli. The CVs could be concatenated to produce nonsense “words” of different lengths. Different number (“n”) of concatenated CVs could be presented at one time from a single loudspeaker located on the azimuth plane or from different loudspeakers. Listeners determined if “n + delta n” sound sources were louder than “n” sound sources. The individual levels of the concatenated CVs was another independent variable. Preliminary data suggest that when “n” is small (<4), the level of the individual sounds determines loudness judgements. When n is larger (n>4) the overall level of the simultaneously presented sounds determines loudness judgements. A full study will be presented. (Work supported by grants from NIDCD and Oculus VR, LLC.)

Evolution of Vocal Diversity through Morphological Adaptation without Vocal Learning or Complex Neural Control

The evolution of complex behavior is driven by theinterplay of morphological specializations and neuromuscular control mechanisms [1-3], and it is often difficult to tease apart their respective contributions. Avian vocal learning and associated neural adaptations are thought to have played a major role in bird diversification [4-8], whereas functional significance of substantial morphological diversity of the vocal organ remains largely unexplored. Within the most species-rich order, Passeriformes, "tracheophones" are a suboscine group that, unlike their oscine sister taxon, does not exhibit vocal learning [9] and is thought to phonate with tracheal membranes [10, 11] instead of the two independent sources found in other passerines [12-14]. Here we show tracheophones possess three sound sources, two oscine-like labial pairs and the unique tracheal membranes, which collectively represent the largest described number of sound sources for a vocal organ. Birds with experimentally disabled tracheal membranes were still able to phonate. Instead of the main sound source, the tracheal membranes constitute a morphological specialization, which, through interaction with bronchial labia, contributes to different acoustic features such as spectral complexity, amplitude modulation, and enhanced sound amplitude. In contrast, these same features arise in oscines from neuromuscular control of two labial sources [15-17]. These findings are supported by a modeling approach andprovide a clear example for how a morphological adaptation of the tracheophone vocal organ cangenerate specific, complex sound features. Morphological specialization therefore constitutes an alternative path in the evolution of acoustic diversity to that of oscine vocal learning and complex neural control.

Measuring speech intelligibility with speech and noise interferers in a loudspeaker-based virtual sound environment

Loudspeaker-based virtual sound environments (VSEs) are emerging as a versatile tool for studying human auditory perception. In order to investigate the reproducibility of simple sound scenes, speech reception thresholds (SRTs) were measured with two interferers and in two spatial conditions (co-located and ±30° separated) using the Danish matrix sentence test Dantale II (Wagener et al., 2003). SRTs were measured in a typical listening room and in a VSE consisting of a spherical 64-channel loudspeaker array using simulated room acoustics with mixed-order-Ambisonics (MOA) playback. The speech maskers were taken from the same material as the target (different talker, same sex). The noise maskers had the same long-term spectrum and broadband envelope as the speech interferer but had random phase (Best et al., 2013). The co-located conditions were reproduced comparably in the real room and in the VSE, with both speech and noise interferers. However, spatial separation led to a 3 dB higher benefit in the VSE than in the real room in both interferer conditions. Previous studies using a larger number of sound sources and more reverberation did not show such systematic differences between virtual and reference conditions, suggesting that reproduction errors may be masked in more complex scenes.

How many images are in an auditory scene?

If an auditory scene consists of many spatially separated sound sources, how many sound sources can be processed by the auditory system? Experiment I determined how many speech sources could be localized simultaneously on the azimuth plane. Different words were played from multiple loudspeakers, and listeners reported the total number of sound sources and their individual locations. In experiment II the accuracy of localizing one speech source in a mixture of multiple speech sources was determined. An extra sound source was added to an existing set of sound sources, and the task was to localize that extra source. In experiment III the setup and task were the same as in experiment I, except that the sounds were tones. The results showed that the maximum number of sound sources that listeners could perceive was limited to approximately four spatially separated speech signals and three for tonal signals. The localization errors increased along with the increase of total number of sound sources. When four or more speech sources already existed, the accuracy in localizing an additional source was near chance.

Arrangement of control points for sound spot generation in the multi-point control method

In this research, we considered control point arrangement by using small-scale linear loudspeaker array. The multi-point control method is proposed for sound reproduction in a limited area/direction. This method is based on boundary surface control principle. The desired signal is reproduced in the target point by the sound pressure control in the control points. According to the MINT (multiple-input/output theorem), the signal is reproduced with accuracy in the target point if the number of sound sources more than the number of control points exists. However, a lot of speakers make it difficult to use in real environment. Therefore, small-scale loudspeaker array is preferable. If the number of control points is greater than the number of sound sources, the filter coefficient is approximated by the least squares-method. This approximation affects the sound pressure suppression and sound quality depending on the arrangement of control point. We examined the number of control points arranged. The input signal is the speech signal that has 300-3400 Hz frequency band. We evaluated some arrangement pattern by sound pressure, SD (Spectral Distortion), and PESQ (Perceptual Evaluation of Speech Quality). As a result, good results were obtained than conventional arrangement.

Sound source localization based on sparse estimation and convex clustering

Sound source localization techniques in array signal processing have been interested for many years. A lot of direction-of-arrival (DOA) estimation methods have been proposed. Most of them use plane waves as model. On the other hand, 3D sound source localization techniques are important for many applications. They are required to estimate the DOA and the distance of sound sources. In the case of the techniques targeted to estimate the distances of sound sources, monopoles must be used for the model. The positions of monopoles are candidates for the sound source positions. However, it is difficult to handle a large number of monopoles placed at various distances. Moreover, it is difficult to decide the number of sound sources to be estimated from the candidates. The proposed method uses sparse estimation with a monopole dictionary in frequency domain, as is well known that the effectiveness of sparse representation for DOA estimation has been in late years. After the sparse estimation, the proposed method ...

Active binaural localization of multiple sound sources

Sound source localization serves as a significant capability of autonomous robots that conduct missions such as search and rescue, and target tracking in challenging environments. However, localization of multiple sound sources and static sound source tracking in self-motion are both difficult tasks, especially when the number of sound sources or reflections increase. This study presents two robotic hearing approaches based on a human perception model (Wallach, 1939) that combines interaural time difference (ITD) and head turn motion data to locate sound sources. The first method uses a fitting-based approach to recognize the changing trends of the cross-correlation function of binaural inputs. The effectiveness of this method was validated using data collected from a two-microphone array rotating in a non-anechoic environment, and the experiments reveal its ability to separate and localize up to three sound sources of the same spectral content (white noise) at different azimuth and elevation angles. The second method uses an extended Kalman filter (EKF) that estimates the orientation of a sound source by fusing the robot’s self-motion and ITD data to reduce the localization errors recursively. This method requires limited memory resources and is able to keep tracking the relative position change of a number of static sources when the robot moves. In the experiments, up to three sources can be tracked simultaneously with a two-microphone array.

De números a números: Visualización del sonido por métodos relacionales de muestreo en directo

Desarrollo de un método de visualización de sonido basado en las muestras numéricas obtenidas como resultado de su digitalización en tiempo real, siendo adecuado para eventos audiovisuales en directo dentro de la tradición de Música Visual. Teniendo en cuenta que: un sonido digitalizado no es más que un conjunto de números ordenados en el tiempo y una imagen no es más que un conjunto de pixeles ordenados en el espacio; el método que proponemos toma los valores numéricos de las muestras de sonido y los ubica espacialmente generando imágenes en movimiento. En función del método usado para la colocación espacial de las muestras, el número muestras de sonido y el número de fuentes sonoras, obtenemos diferentes familias de algoritmos de visualización con características estéticas propias. Este trabajo presenta las bases necesarias para abordar la visualización del sonido basada en muestras y presenta casos de estudio realizados con una aplicación desarrollada en el lenguaje de programación Java.Palabras-clave: AUDIOVISUALIZACIÓN; MÚSICA VISUAL; VISUALIZACIÓN DE DATOS; TIEMPO REAL; SONIDO EXPERIMENTAL Numbers to numbers: Audiovisualization for relational live sampling methods AbstractDevelopment of a sound visualization method based on digital samples, these samples were obtained from the real-time scanning. The method is suitable for audiovisual live events within the tradition of Visual Music. If we consider that: a digitized sound is just a set of numbers arranged in time, and an image it is no more than a set of pixels ordained in space. The proposed method takes the numerical values of the sound samples and spatially located they, generating a moving image. Depending on the method used for the spatial positioning of the samples, the number of sound samples used, and the number of sound sources, we get different visualization algorithms families with own aesthetic characteristics. This work establishes the necessary measures to address the sound visualization based on samples, and presents case studies performed with a sofware developed in Java programming language.Keywords: AUDIOVISUALIZATION; VISUAL MUSIC; DATA VISUALIZATION; REAL TIME; EXPERIMENTAL SOUND

Linear multivariate evaluation models for spatial perception of soundscape.

Soundscape is a sound environment that emphasizes the awareness of auditory perception and social or cultural understandings. The case of spatial perception is significant to soundscape. However, previous studies on the auditory spatial perception of the soundscape environment have been limited. Based on 21 native binaural-recorded soundscape samples and a set of auditory experiments for subjective spatial perception (SSP), a study of the analysis among semantic parameters, the inter-aural-cross-correlation coefficient (IACC), A-weighted-equal sound-pressure-level (L(eq)), dynamic (D), and SSP is introduced to verify the independent effect of each parameter and to re-determine some of their possible relationships. The results show that the more noisiness the audience perceived, the worse spatial awareness they received, while the closer and more directional the sound source image variations, dynamics, and numbers of sound sources in the soundscape are, the better the spatial awareness would be. Thus, the sensations of roughness, sound intensity, transient dynamic, and the values of Leq and IACC have a suitable range for better spatial perception. A better spatial awareness seems to promote the preference slightly for the audience. Finally, setting SSPs as functions of the semantic parameters and Leq-D-IACC, two linear multivariate evaluation models of subjective spatial perception are proposed.

Audio Fingerprinting for Multi-Device Self-Localization

We investigate the self-localization problem of an ad-hoc network of randomly distributed and independent devices in an open-space environment with low reverberation but heavy noise (e.g. smartphones recording videos of an outdoor event). Assuming a sufficient number of sound sources, we estimate the distance between a pair of devices from the extreme (minimum and maximum) time difference of arrivals (TDOAs) from the sources to the pair of devices without knowing the time offset. The obtained inter-device distances are then exploited to derive the geometrical configuration of the network. In particular, we propose a robust audio fingerprinting algorithm for noisy recordings and perform landmark matching to construct a histogram of the TDOAs of multiple sources. The extreme TDOAs can be estimated from this histogram. By using audio fingerprinting features, the proposed algorithm works robustly in very noisy environments. Experiments with free-field simulation and open-space recordings prove the effectiveness of the proposed algorithm.

Direction-of-arrival estimations based on a two-microphone array using two levels of Bayesian inference

Due to a wide variety of potential applications, such as in video conferencing, mobile devices, and robotics, sound source localization using a small aperture consisting of only two microphones has been actively investigated. Based on the observed time-differences of arrival between sound signals, a probability distribution of the direction of the sources is derived to estimate the actual direction of sources. Many existing algorithms, however, assume a given number of sound sources. This paper describes a recent development in a model-based Bayesian probabilistic approach by Escolano et al. [J. Acoust. Am. Soc. 135, 742–751 (2014)], which allows both the number and direction of speech sources to be inferred. This paper will demonstrate that a unified framework encompassing two levels of Bayesian inference, model selection and parameter estimation, can be efficiently applied in this challenging task. This paper will also present different experimental setups and scenarios to investigate the performance of the proposed method.

Multiple sound source localization when sounds are stationary or rotating

In the field of spatial hearing, it is known that the sound source localization error increases when distractors are present. This study first determined the maximum number of sound sources that can be perceived. In one experiment, listeners determined how many sources (loudspeakers) presented sounds and these loudspeaker locations. In another experiment, listeners were asked to locate a new sound source which was added to an existing set of fixed sources. Listeners determined up to about four sources producing simultaneous unrelated speech sounds in both experiments. A second part of the research examined the effect of rotating the sound sources on the listener’s perception. If multiple unrelated sounds such as speech rotated, listeners could much more easily determine sound rotation than if the sounds were related to each other. If the sounds were related (e.g., all sounds were harmonics of a fundamental) but were still presented from a number of different loudspeakers, listeners could hardly process the individual sounds even when sounds rotated among the loudspeakers. Thus, the ability to locate multiple sound sources depends on the total number and motion of sources and the type of sound. [Research supported by an AFOSR grant.]

Parametric Spatial Sound Processing: A flexible and efficient solution to sound scene acquisition, modification, and reproduction

Flexible and efficient spatial sound acquisition and subsequent processing are of paramount importance in communication and assisted listening devices such as mobile phones, hearing aids, smart TVs, and emerging wearable devices (e.g., smart watches and glasses). In application scenarios where the number of sound sources quickly varies, sources move, and nonstationary noise and reverberation are commonly encountered, it remains a challenge to capture sounds in such a way that they can be reproduced with a high and invariable sound quality. In addition, the objective in terms of what needs to be captured, and how it should be reproduced, depends on the application and on the user?s preferences. Parametric spatial sound processing has been around for two decades and provides a flexible and efficient solution to capture, code, and transmit, as well as manipulate and reproduce spatial sounds.

Performance of active feedforward control systems in non-ideal, synthesized diffuse sound fields.

The acoustic performance of passive or active panel structures is usually tested in sound transmission loss facilities. A reverberant sending room, equipped with one or a number of independent sound sources, is used to generate a diffuse sound field excitation which acts as a disturbance source on the structure under investigation. The spatial correlation and coherence of such a synthesized non-ideal diffuse-sound-field excitation, however, might deviate significantly from the ideal case. This has consequences for the operation of an active feedforward control system which heavily relies on the acquisition of coherent disturbance source information. This work, therefore, evaluates the spatial correlation and coherence of ideal and non-ideal diffuse sound fields and considers the implications on the performance of a feedforward control system. The system under consideration is an aircraft-typical double panel system, equipped with an active sidewall panel (lining), which is realized in a transmission loss facility. Experimental results for different numbers of sound sources in the reverberation room are compared to simulation results of a comparable generic double panel system excited by an ideal diffuse sound field. It is shown that the number of statistically independent noise sources acting on the primary structure of the double panel system depends not only on the type of diffuse sound field but also on the sample lengths of the processed signals. The experimental results show that the number of reference sensors required for a defined control performance exhibits an inverse relationship to control filter length.

A Bayesian direction-of-arrival model for an undetermined number of sources using a two-microphone array.

Sound source localization using a two-microphone array is an active area of research, with considerable potential for use with video conferencing, mobile devices, and robotics. Based on the observed time-differences of arrival between sound signals, a probability distribution of the location of the sources is considered to estimate the actual source positions. However, these algorithms assume a given number of sound sources. This paper describes an updated research account on the solution presented in Escolano et al. [J. Acoust. Am. Soc. 132(3), 1257-1260 (2012)], where nested sampling is used to explore a probability distribution of the source position using a Laplacian mixture model, which allows both the number and position of speech sources to be inferred. This paper presents different experimental setups and scenarios to demonstrate the viability of the proposed method, which is compared with some of the most popular sampling methods, demonstrating that nested sampling is an accurate tool for speech localization.

Combined effects of auditory and visual cues on the perception of vection

Vection is the illusion of self-motion in the absence of real physical movement. The aim of the present study was to analyze how multisensory inputs (visual and auditory) contribute to the perception of vection. Participants were seated in a stationary position in front of a large, curved projection display and were exposed to a virtual scene that constantly rotated around the yaw-axis, simulating a 360° rotation. The virtual scene contained either only visual, only auditory, or a combination of visual and auditory cues. Additionally, simulated rotation speed (90°/s vs. 60°/s) and the number of sound sources (1 vs. 3) were varied for all three stimulus conditions. All participants were exposed to every condition in a randomized order. Data specific to vection latency, vection strength, the severity of motion sickness (MS), and postural steadiness were collected. Results revealed reduced vection onset latencies and increased vection strength when auditory cues were added to the visual stimuli, whereas MS and postural steadiness were not affected by the presence of auditory cues. Half of the participants reported experiencing auditorily induced vection, although the sensation was rather weak and less robust than visually induced vection. Results demonstrate that the combination of visual and auditory cues can enhance the sensation of vection.

A speech distortion weighting based approach to integrated active noise control and noise reduction in hearing aids

This paper presents weighted approaches for integrated active noise control and noise reduction in hearing aids. The unweighted integrated active noise control and noise reduction scheme introduced in the previous work does not allow to trade-off between the active noise control and the noise reduction. In some circumstances it will, however, be useful to emphasize one of the functional blocks.Changing the original optimisation problem to a constrained optimisation problem leads to a scheme based on a weighted mean squared error criterion that allows to focus either on the active noise control or on the noise reduction. It is similarly possible to derive a scheme that allows to focus either on reducing the speech distortion or on reducing the residual noise at the eardrum. In a single speech source scenario and when the number of sound sources (speech plus noise sources) is less than or equal to the number of microphones, it is possible to derive a simple formula for the output signal-to-noise ratio of the latter scheme. It can then be shown that this scheme delivers a constant signal-to-noise ratio at the eardrum for any weighting factor.

Feature assignment in perception of auditory figure.

Because the environment often includes multiple sounds that overlap in time, listeners must segregate a sound of interest (the auditory figure) from other co-occurring sounds (the unattended auditory ground). We conducted a series of experiments to clarify the principles governing the extraction of auditory figures. We distinguish between auditory "objects" (relatively punctate events, such as a dog's bark) and auditory "streams" (sounds involving a pattern over time, such as a galloping rhythm). In Experiments 1 and 2, on each trial 2 sounds-an object (a vowel) and a stream (a series of tones)-were presented with 1 target feature that could be perceptually grouped with either source. In each block of these experiments, listeners were required to attend to 1 of the 2 sounds, and report its perceived category. Across several experimental manipulations, listeners were more likely to allocate the feature to an impoverished object if the result of the grouping was a good, identifiable object. Perception of objects was quite sensitive to feature variation (noise masking), whereas perception of streams was more robust to feature variation. In Experiment 3, the number of sound sources competing for the feature was increased to 3. This produced a shift toward relying more on spatial cues than on the potential contribution of the feature to an object's perceptual quality. The results support a distinction between auditory objects and streams, and provide new information about the way that the auditory world is parsed.

소리의 공간 제어를 위한 구형 다채널 스피커 어레이 설계

The objective of this paper is to design multichannel spherical loudspeaker array by considering various positioning methods such as Gaussian grid, Lebedev grid and packing method. For the spatial sound manipulation, which is to make desired sound field by controling multiple sound sources, the Kirchhoff- Helmholtz integral states that sound fields can be reproduced in terms of infinite control sources on the integral surface. But since we cannot control infinite number of sources for the implementation, we have to allocate finite number of sound sources which can approximately act as infinite number of sources. To manipulate sound field inside of a sphere (which is typical example of three dimensional array) by controlling sound sources on the surface, three methods of allocating sound sources, which are Gaussian grid, Lebedev grid and packing method, are reviewed. For each geometry, the performances of manipulation rendered by time-reversal operator and higher-order ambisonics are compared.

A Nonparametric Bayesian Multipitch Analyzer Based on Infinite Latent Harmonic Allocation

The statistical multipitch analyzer described in this paper estimates multiple fundamental frequencies (F0s) in polyphonic music audio signals produced by pitched instruments. It is based on hierarchic4al nonparametric Bayesian models that can deal with uncertainty of unknown random variables such as model complexities (e.g., the number of F0s and the number of harmonic partials), model parameters (e.g., the values of F0s and the relative weights of harmonic partials), and hyperparameters (i.e., prior knowledge on complexities and parameters). Using these models, we propose a statistical method called infinite latent harmonic allocation (iLHA). To avoid model-complexity control, we allow the observed spectra to contain an unbounded number of sound sources (F0s), each of which is allowed to contain an unbounded number of harmonic partials. More specifically, to model a set of time-sliced spectra, we formulated nested infinite Gaussian mixture models based on hierarchical and generalized Dirichlet processes. To avoid manual tuning of influential hyperparameters, we put noninformative hyperprior distributions on them in a hierarchical manner. For efficient Bayesian inference, we used a modern technique called collapsed variational Bayes. In comparative experiments using audio recordings of piano and guitar solo performances, iLHA yielded promising results and we found that there would be room for improvement based on modeling of temporal continuity and spectral smoothness.