Binaural sound source localization based on generalized parametric model and two-layer matching strategy in complex environments

Abstract

Binaural sound source localization is an important technique involving Human-Robot Interaction (HRI), video conference, speech enhancement, etc. In many real application scenarios, especially for closed environments, the affect of reverberation and noise would degrade the precision of position estimations. Therefore, a new binaural sound source localization method based on generalized parametric model and two-layer matching strategy is proposed in this paper for complex environments. Firstly, cepstral prefiltering is utilized for dereverberation of binaural signals. Then, two binaural cues computed from a dual-channel frequency representation, are combined to estimate the azimuths of sources. Additionally, the generalized parametric model is presented to describe the relationship between the azimuth and binaural cues through finding the optimal scaling factors from training data. At last, a two-layer matching strategy based on Bayesian rule is used to make the final decision, which can effectively decrease the computation complexity. Experiments have validated the proposed approach and show that it achieves favorably better results compared with several available methods without extra spacial burden.

Binaural Sound Localization Based on Reverberation Weighting and Generalized Parametric Mapping

“…The first row of panels is only based on ITDs through Eq. (29). It can be seen that the ITD-based azimuth estimation becomes more and more ambiguous with increasing frequency.…”

“…Moreover, the binaural signals collected by an artificial head in a realistic indoor room, are also used to validate the proposed method in practical situations. Note that the methods proposed in this paper are the refined and expanded version of the conference proceedings papers [28] and [29].…”

“…The averaged ITD and IID templates are established using the training data, namely HRTFs. Then, the generalized parametric mapping is proposed to build a generalized parametric model [29] through finding generalized scaling factors. This model describes the mapping relationship from the extracted binaural cues to azimuth estimates.…”

“…Through suppressing the early and late reverberation separately, the proposed reverberation weighting can preserve binaural cues better. Although cepstral prefiltering is used to dereverberate binaural signals for localization [29], [30], it mainly focuses on the dereverberation for time-delay estimation [31].…”

“…It can also improve the adaptability of the proposed method to practical applications. The generalized parametric model in [29] is improved by parametric mapping with the optimal generalized scaling factors which are obtained by solving a least squares problem. Although the joint estimation based on ITD and IID has been studied in [24], how to find more generalized scaling factors that are used to describe the mapping relationship from the extracted binaural cues to the objective azimuth, has not been provided.…”