Conference Paper

Blind alignment of asynchronously recorded signals for distributed microphone array

Grad. Sch. of Inf. Sci. & Technol., Univ. of Tokyo, Tokyo, Japan
DOI: 10.1109/ASPAA.2009.5346505 Conference: Applications of Signal Processing to Audio and Acoustics, 2009. WASPAA '09. IEEE Workshop on
Source: IEEE Xplore


In this paper, aiming to utilize independent recording devices as a distributed microphone array, we present a novel method for alignment of recorded signals with localizing microphones and sources. Unlike conventional microphone array, signals recorded by independent devices have different origins of time, and microphone positions are generally unknown. In order to estimate both of them from only recorded signals, time differences between channels for each source are detected, which still include the differences of time origins, and an objective function defined by their square errors is minimized. For that, simple iterative update rules are derived through auxiliary function approach. The validity of our approach is evaluated by simulative experiment.

Download full-text


Available from: Shigeki Sagayama
  • Source
    • "In this experiment the TDOA of each source with respect to a pair of microphones is estimated from the corresponding segmentations in the microphone signals. A coarse-to-fine scheme[5]is employed to estimate the TDOA, where the two microphone signals are coarsely aligned at first and then processed with the generalized cross-correlation with phase transform (GCC-PHAT) algorithm[39], which is well-known for its robustness to room reverberation. In the GCC-PHAT algorithm, we use a frame length of 8192 with half overlap. "
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigate the problem of sensor and source joint localization using time-difference of arrivals (TDOAs) of an ad-hoc array. A major challenge is that the TDOAs contain unknown time offsets between asynchronous sensors. To address this problem, we propose a low-rank approximation method that does not need any prior knowledge of sensor and source locations or timing information. At first, we construct a pseudo time of arrival (TOA) matrix by introducing two sets of unknown timing parameters (source onset times and device capture times) into the current TDOA matrix. Then we propose a Gauss-Newton low-rank approximation algorithm to jointly identify the two sets of unknown timing parameters, exploiting the low-rank property embedded in the pseudo TOA matrix. We derive the boundaries of the timing parameters to reduce the initialization space and employ a multi-initialization scheme. Finally, we use the estimated timing parameters to correct the pseudo TOA matrix, which is further applied to sensor and source localization. Experimental results show that the proposed approach outperforms state-of-the-art algorithms.
    Full-text · Article · Feb 2016 · IEEE Transactions on Signal Processing
  • Source
    • "Nonetheless, better localization of these parameters is dependent on the precision of the estimated time difference of arrivals (TDOAs) between the direct sound and the reflected sounds. Even though the current method of estimating the TDOAs involves the use of multiple sensors [3] [4] [5], advance synchronization among sensors is of critical importance [6]. Recent approaches are, thereby, geared toward the utilization of few sensors in TDOAs estimation, which is advantageous for ease of application. "

    Preview · Article · Jan 2016 · Acoustical Science and Technology
  • Source
    • "A challenge in such ad-hoc arrays is that the locations of the microphones are generally unknown and there is no precise temporal synchronization between the microphones . Traditional microphone-array techniques, such as beamforming and sound source localization, which rely on the knowledge of microphone positions and assume samplesynchronized audio channels, cannot be applied directly [2] [3]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: We use audio fingerprinting to solve the synchronization problem between multiple recordings from an ad-hoc array consisting of randomly placed wireless microphones or hand-held smartphones. Synchronization is crucial when employing conventional microphone array techniques such as beamforming and source localization. We propose a fine audio landmark fingerprinting method that detects the time difference of arrivals (TDOAs) of multiple sources in the acoustic environment. By estimating the maximum and minimum TDOAs, the proposed method can accurately calculate the unknown time offset between a pair of microphone recordings. Experimental results demonstrate that the proposed method significantly improves the synchronization accuracy of conventional audio fingerprinting methods and achieves comparable performance to the generalized cross-correlation method.
    Full-text · Conference Paper · Sep 2015
Show more