Sound Source Localization Method Research Articles

Moving sound source localization based on triangulation method

This study develops a sound source localization method that extends traditional triangulation to moving sources. First, the possible sound source locating plane is scanned. Secondly, for each hypothetical source location in this possible plane, the Doppler effect is removed through the integration of sound pressure. Taking advantage of the de-Dopplerized signals, the moving time difference of arrival (MTDOA) is calculated, and the sound source is located based on triangulation. Thirdly, the estimated sound source location is compared to the original hypothetical location and the deviations are recorded. Because the real sound source location leads to zero deviation, the sound source can be finally located by minimizing the deviation matrix. Simulations have shown the superiority of MTDOA method over traditional triangulation in case of moving sound sources. The MTDOA method can be used to locate moving sound sources with as high resolution as DAMAS beamforming, as shown in the experiments, offering thus a new method for locating moving sound sources.

Real-time multiple sound source localization and counting using a soundfield microphone

In this work, a multiple sound source localization and counting method based on a relaxed sparsity of speech signal is presented. A soundfield microphone is adopted to overcome the redundancy and complexity of microphone array in this paper. After establishing an effective measure, the relaxed sparsity of speech signals is investigated. According to this relaxed sparsity, we can obtain an extensive assumption that “single-source” zones always exist among the soundfield microphone signals, which is validated by statistical analysis. Based on “single-source” zone detecting, the proposed method jointly estimates the number of active sources and their corresponding DOAs by applying a peak searching approach to the normalized histogram of estimated DOA. The cross distortions caused by multiple simultaneously occurring sources are solved by estimating DOA in these “single-source” zones. The evaluations reveal that the proposed method achieves a higher accuracy of DOA estimation and source counting compared with the existing techniques. Furthermore, the proposed method has higher efficiency and lower complexity, which makes it suitable for real-time applications.

Estimation of seismic wave velocity at seafloor surface and sound source localization based on transmitted wave observation with an ocean bottom seismometer offshore of Kamaishi, Japan

An in situ method of estimating the seismic wave velocity at the seafloor surface by observing the particle motion of a wave transmitted into the sediment is presented; this method uses a sound source whose location is known. Conversely, a sound source localization method using the obtained seismic velocities and involving particle motion observation is also presented. Although this method is applicable only when the sound source exists within the critical incidence angle range, it is expected to contribute to the tracing of vocalizing baleen whales, which are unknown around Japanese waters.

Development of the Source Reconstruction System by Combining Sound Source Localization and Time Reversal Method

AbstractThis study explored the target sound source location at unknown situation and processed the received signal to determine the location of the target, including the reconstructed signal of source immediately. In this paper, it used triangulation sound sources localization and time reversal method (TRM) to reconstruct the source signals. The purpose is to use a sound source localization method with a simple device to quickly locate the position of the sound source. This method uses the microphone array to measure signal from the target sound source. Then, the sound source location is calculated and is indicated by Cartesian coordinates. The sound source location is then used to evaluate free field impulse response function which can replace the impulse response function used in time-reversal method. This process reduces the computation time greatly which makes possible for a real time source localization and source signal separation.

Research of HRTF Character of Head-Mounted Microphone Array

Head-mounted microphone array has practical applications in robot acoustic localization system and wearable anti-sniper positioning system. Usually, sound source localization methods are based on linear or nonlinear unblocked microphone arrays. But head-mounted microphone array is a kind of blocked arrays, with which it needs information of Head Related Transfer Function (HRTF) for precise localization. In this paper, we research the HRTF character of head-mounted microphone array for localization in high frequency band and low frequency band respectively, and design a localization algorithm for low frequency sound based on head-mounted microphone array to analysis the threshold between high and low frequency. Experimental results show that the Head-mounted Microphone Array causes diffraction effect for low frequency sound, and amplitude attenuation effect for high frequency sound, and when the low frequency band is limited into 1 KHz, the localization algorithm for low frequency realizes the best performance.

Speaker localization in noisy environments using steered response voice power

Many devices, including smart TVs and humanoid robots, can be operated through speech interface. Since a user can interact with such a device at a distance, speech-operated devices must be able to process speech signals from a distance. Although many methods exist to localize speakers via sound source localization, it is very difficult to reliably find the location of a speaker in a noisy environment. In particular, conventional sound source localization methods only find the loudest sound source within a given area, and such a sound source may not necessarily be related to human speech. This can be problematic in real environments where loud noises frequently occur, and the performance of speech-based interfaces for a variety of devices could be negatively impacted as a result. In this paper, a new speaker localization method is proposed. It identifies the location associated with the maximum voice power from all candidate locations. The proposed method is tested under a variety of conditions using both simulation data and real data, and the results indicate that the performance of the proposed method is superior to that of a conventional algorithm for various types of noises1.

MUSICに基づくパーティクルフィルタによる２音源追尾の高速化

In this paper, MUSIC (multiple signal classification) algorithm is applied for a sound source tracking using particle filters. MUSIC algorithm is a high-resolution sound source localization method but has high computational costs. Therefore, a reduction of computational cost is attempted by using the MUSIC spectrum function as a likelihood function of particle filters. When there are two sound sources, however, a particle intensification problem arises simultaneously. Therefore, a method to overcome this problem is also proposed. Several experimental results in a reverberant environment show that the proposed method can track two sound sources in real time.

A Steered-Response Power Algorithm Employing Hierarchical Search for Acoustic Source Localization Using Microphone Arrays

The localization of a speaker inside a closed environment is often approached by real-time processing of multiple audio signals captured by a set of microphones. One of the leading related methods for sound source localization, the steered-response power (SRP), searches for the point of maximum power over a spatial grid. High-accuracy localization calls for a dense grid and/or many microphones, which tends to impractically increase computational requirements. This paper proposes a new method for sound source localization (called H-SRP), which applies the SRP approach to space regions instead of grid points. This arrangement makes room for the use of a hierarchical search inspired by the branch-and-bound paradigm, which is guaranteed to find the global maximum in anechoic environments and shown experimentally to also work under reverberant conditions. Besides benefiting from the improved robustness of volume-wise search over point-wise search as to reverberation effects, the H-SRP attains high performance with manageable complexity. In particular, an experiment using a 16-microphone array in a typical presentation room yielded localization errors of the order of 7 cm, and for a given fixed complexity, competing methods' errors are two to three times larger.

Improved sound source localization in horizontal plane for binaural robot audition

An improved sound source localization (SSL) method has been developed that is based on the generalized cross-correlation (GCC) method weighted by the phase transform (PHAT) for use with binaural robots equipped with two microphones inside artificial pinnae. The conventional SSL method based on the GCC-PHAT method has two main problems when used on a binaural robot platform: 1) diffraction of sound waves with multipath interference caused by the contours of the robot head, which affects localization accuracy, and 2) front-back ambiguity, which limits the localization range to half the horizontal space. The diffraction problem was overcome by incorporating a new time delay factor into the GCC-PHAT method under the assumption of a spherical robot head. The ambiguity problem was overcome by utilizing the amplification effect of the pinnae for localization over the entire azimuth. Experiments conducted using two dummy heads equipped with small or large pinnae showed that localization errors were reduced by 8.91° (3.21° vs. 12.12°) on average with the new time delay factor compared with the conventional GCC-PHAT method and that the success rate for front-back disambiguation using the pinnae amplification effect was 29.76 % (93.46 % vs. 72.02 %) better on average over the entire azimuth than with a conventional head related transfer function (HRTF)-based method.

Evaluation methods for sound source localization systems of speakers in a meeting room

The employing of sound source localization methods allows to evaluate location and head orientation of a speaker in a room. At present such systems are popular at development of intelligent support systems for smart meeting rooms. In this paper, a set of the metrics for performance evaluation of sound source localization systems as well as their integration with video monitoring systems are analyzed. Accuracy estimation of speaker positions, located on 32 chairs was carried out in the developed smart meeting room

Real-Time Multiple Sound Source Localization and Counting Using a Circular Microphone Array

In this work, a multiple sound source localization and counting method is presented, that imposes relaxed sparsity constraints on the source signals. A uniform circular microphone array is used to overcome the ambiguities of linear arrays, however the underlying concepts (sparse component analysis and matching pursuit-based operation on the histogram of estimates) are applicable to any microphone array topology. Our method is based on detecting time-frequency (TF) zones where one source is dominant over the others. Using appropriately selected TF components in these “single-source” zones, the proposed method jointly estimates the number of active sources and their corresponding directions of arrival (DOAs) by applying a matching pursuit-based approach to the histogram of DOA estimates. The method is shown to have excellent performance for DOA estimation and source counting, and to be highly suitable for real-time applications due to its low complexity. Through simulations (in various signal-to-noise ratio conditions and reverberant environments) and real environment experiments, we indicate that our method outperforms other state-of-the-art DOA and source counting methods in terms of accuracy, while being significantly more efficient in terms of computational complexity.

A real-time super-resolution robot audition system that improves the robustness of simultaneous speech recognition

This study addresses a framework for a robot audition system, including sound source localization (SSL) and sound source separation (SSS), that can robustly recognize simultaneous speeches in a real environment. Because SSL estimates not only the location of speakers but also the number of speakers, such a robust framework is essential for simultaneous speech recognition. Moreover, improvement in the performance of SSS is crucial for simultaneous speech recognition because the robot has to recognize the individual source of speeches. For simultaneous speech recognition, current robot audition systems mainly require noise-robustness, high resolution, and real-time implementation. Multiple signal classification (MUSIC) based on standard Eigenvalue decomposition (SEVD) and Geometric-constrained high-order decorrelation-based source separation (GHDSS) are techniques utilizing microphone array processing, which are used for SSL and SSS, respectively. To enhance SSL robustness against noise while detecting simultaneous speeches, we improved SEVD-MUSIC by incorporating generalized Eigenvalue decomposition (GEVD). However, GEVD-based MUSIC (GEVD-MUSIC) and GHDSS mainly have two issues: (1) the resolution of pre-measured transfer functions (TFs) determines the resolution of SSL and SSS and (2) their computational cost is expensive for real-time processing. For the first issue, we propose a TF-interpolation method integrating time-domain-based and frequency-domain-based interpolation. The interpolation achieves super-resolution robot audition, which has a higher resolution than that of the pre-measured TFs. For the second issue, we propose two methods for SSL: MUSIC based on generalized singular value decomposition (GSVD-MUSIC) and hierarchical SSL (H-SSL). GSVD-MUSIC drastically reduces the computational cost while maintaining noise-robustness for localization. In addition, H-SSL reduces the computational cost by introducing a hierarchical search algorithm instead of using a greedy search for localization. These techniques are integrated into a robot audition system using a robot-embedded microphone array. The preliminary experiments for each technique showed the following: (1) The proposed interpolation achieved approximately 1-degree resolution although the TFs are only at 30-degree intervals in both SSL and SSS; (2) GSVD-MUSIC attained 46.4 and 40.6% of the computational cost compared to that of SEVD-MUSIC and GEVD-MUSIC, respectively; (3) H-SSL reduced 71.7% of the computational cost to localize a single speaker. Finally, the robot audition system, including super-resolution SSL and SSS, is applied to robustly recognize four sources of speech occurring simultaneously in a real environment. The proposed system showed considerable performance improvements of up to 7% for the average word correct rate during simultaneous speech recognition, especially when the TFs were at more than 30-degree intervals.

Improved binaural sound localization and tracking for unknown time-varying number of speakers

A method based on the generalized cross-correlation (GCC) method weighted by the phase transform (PHAT) has been developed for binaural sound source localization (SSL) and tracking of multiple sound sources. Accurate binaural audition is important for applying inexpensive and widely applicable auditory capabilities to robots and systems. Conventional SSL based on the GCC-PHAT method is degraded by low resolution of the time difference of arrival estimation, by the interference created when the sound waves arrive at a microphone from two directions around the robot head, and by impaired performance when there are multiple speakers. The low-resolution problem is solved by using a maximum-likelihood-based SSL method in the frequency domain. The multipath interference problem is avoided by incorporating a new time delay factor into the GCC-PHAT method with assuming a spherical robot head. The performance when there are multiple speakers was improved by using a multisource speech tracking method consisting of voice activity detection (VAD) and K-means clustering. The standard K-means clustering algorithm was extended to enable tracking of an unknown time-varying number of speakers by adding two additional steps that increase the number of clusters automatically and eliminate clusters containing incorrect direction estimations. Experiments conducted on the SIG-2 humanoid robot show that this method outperforms the conventional SSL method; it reduces localization errors by 18.1° on average and by over 37° in the side directions. It also tracks multiple speakers in real time with tracking errors below 4.35°.

Accelerated steered response power method for sound source localization via clustering search

The steered response power-phase transform (SRP-PHAT) sound source localization algorithm is robust in a real environment. However, the large computation complexity limits the practical application of SRP-PHAT. For a microphone array, each location corresponds to a set of time differences of arrival (TDOAs), and this paper collects them into a TDOA vector. Since the TDOA vectors in the adjacent regions are similar, we present a fast algorithm based on clustering search to reduce the computation complexity of SRP-PHAT. In the training stage, the K-means or Iterative Self-Organizing Data Analysis Technique (ISODATA) clustering algorithm is used to find the centroid in each cluster with similar TDOA vectors. In the procedure of sound localization, the optimal cluster is found by comparing the steered response powers (SRPs) of all centroids. The SRPs of all candidate locations in the optimal cluster are compared to localize the sound source. Experiments both in simulation environments and real environments have been performed to compare the localization accuracy and computational load of the proposed method with those of the conventional SRP-PHAT algorithm. The results show that the proposed method is able to reduce the computational load drastically and maintains almost the same localization accuracy and robustness as those of the conventional SRP-PHAT algorithm. The difference in localization performance brought by different clustering algorithms used in the training stage is trivial.

An Acoustic Source Localization Method Based on Equal Distances Multi-Sensors Array

The traditional acoustic source is sensitive to time. A novel sound source location method using linear intersection spacing multi-sensors array is provided in this paper. Each array is composed of three spaced nodes, and least squares method is used to calculate the final position according to ternary array results. Multi-arrays method is more robust than the ternary one, and much wider scope is covered. Location scope extends from 120m to 800m when the relative positioning error is 10%. A multi-array group based on linear intersection sound source localization method is provided in this paper too. Experiment results show that the proposed method has higher precision on angle locating than distance locating.

Applying direct algebraic sound source localization method for time-domain reflectometry of conference room

A novel localization method has been previously exposed, using an explicit formula of direction and distance of a monopole source and a circular array [S. Ando, ASA Seattle Meeting, 2011]. However, this localization method has little been applied in real environments, such as concert halls or conference rooms. It has also been shown that the algorithm has a high temporal resolution [T.Levy, S. Ando, Hong Kong Acoustics 2012, 2012] that enables to localize sound sources using reflected sound waves. Thus, the aim of this study is to study conference room’s reflection using the proposed algorithm and to test its robustness and its efficiency in such conditions. In the experiments, the main reflectors in the conference room will be identified and a comparison between the previously proposed method and the traditional sound source localization algorithms is done, in terms of rapidness and precision.

Real-Time Sound Source Localization for a Mobile Robot Based on the Guided Spectral-Temporal Position Method

Sound source localization based on a microphone array for mobile robots faces great challenges because of factors such as uncertainty of the robot's movement, noise and reverberation, the requirements of a compact microphone array and so on. This paper studies a method for sound source localization in a dynamic environment using a cross-microphone array plane, which consists of four microphones on an intelligent mobile robot platform. Firstly, the method of spectral subtraction and campestral mean normalization is introduced to de-noise. Then GCC-PHAT-ργ and the guided spectral-temporal (ST) position method are proposed to suppress noise and reverberation based on the generalized cross-correlation method to estimate time delay. Finally, the sound source is positioned by adopting the geometric location method. This system is tested by a total of 2,016 sets of experiments. Even in an intensely noisy and reverberating environment, the guided ST position method achieves angle positioning accuracy of more than 95% with a less than 15 degrees localization error. Meanwhile, all the experimental data can be processed in real-time, within 0.4s.

507 音源探査法を用いたリニアスライダの騒音解析

507 音源探査法を用いたリニアスライダの騒音解析

Improved steered response power method for sound source localization based on principal eigenvector

Sound source localization is essential in many microphone arrays application, ranging from teleconferencing systems to artificial perception in a reverberant noisy environment. The steered response power (SRP) using the phase transform (SRP-PHAT) source localization algorithm has been proved robust, however, the performance of the SRP-PHAT algorithm degrades in highly reverberant noisy environment. Though the SRP-based maximum likelihood localizers are more robust than SRP-PHAT, they have the drawback of requiring noise variance to be estimated in a silent room. This paper presents an improved SRP-PHAT algorithm based on principal eigenvector. Sound source location is estimated from the principal eigenvector computed from the frequency-domain correlation matrix. Using both simulated and real data, we show that the proposed algorithm achieves higher source localization accuracy compared to the SRP-PHAT algorithm.

Sound source localization of digital hearing aids using wavelet based multivariate statistical method

The letter proposed a sound source localization method of digital hearing aids using wavelet based multivariate statistics with the Generalized Cross Correlation (GCC) algorithm. Haar wavelet is used to decompose GCC sequences and extract four wavelet characteristics. And then, Hotelling T2 statistical method is used to fuse the four wavelet characteristics. The statistical value is used to judge the number of sound sources and obtain corresponding time delay estimation which is used to localize the position of sound source. The experimental results show that the proposed method has better robustness in an environment with severe noise and reverberation. Meanwhile, the complexity of algorithm is moderate, which is available for sound source localization of hearing aids.