Intuitive spoken dialogues are a prerequisite for human-robot interaction. In many practical situations, robots must be able to identify and focus on sources of interest in the presence of interfering speakers. Techniques such as spatial filtering and blind source separation are therefore often used, but rely on accurate knowledge of the source location. In practice, sound emitted in enclosed environments is subject to reverberation and noise. Hence, sound source localization must be robust to both diffuse noise due to late reverberation, as well as spurious detections due to early reflections. For improved robustness against reverberation, this paper proposes a novel approach for sound source tracking that constructively exploits the spatial diversity of a microphone array installed in a moving robot. In previous work, we developed speaker localization approaches using expectation-maximization (EM) approaches and using Bayesian approaches. In this paper we propose to combine the EM and Bayesian approach in one framework for improved robustness against reverberation and noise.

This paper proposes a novel audio-visual tracking approach that exploits constructively audio and visual modalities in order to estimate trajectories of multiple people in a joint state space. The tracking problem is modeled using a sequential Bayesian filtering framework. Within this framework, we propose to represent the posterior density with a Gaussian Mixture Model (GMM). To ensure that a GMM representation can be retained sequentially over time, the predictive density is approximated by a GMM using the Unscented Transform. While a density interpolation technique is introduced to obtain a continuous representation of the observation likelihood, which is also a GMM. Furthermore, to prevent the number of mixtures from growing exponentially over time, a density approximation based on the Expectation Maximization (EM) algorithm is applied, resulting in a compact GMM representation of the posterior density. Recordings using a camcorder and microphone array are used to evaluate the proposed approach, demonstrating significant improvements in tracking performance of the proposed audio-visual approach compared to two benchmark visual trackers.

Accurate estimation of the Direction of Arrival (DOA) of a sound source is an important prerequisite for a wide range of acoustic signal processing applications. However, in enclosed environments, early reflections and late reverberation often lead to localization errors. Recent work demonstrated that improved robustness against reverberation can be achieved by clustering only the DOAs from direct-path bins in the short-term Fourier transform of a speech signal of several seconds duration from a static talker. Nevertheless, for moving talkers, short blocks of at most several hundred milliseconds are required to capture the spatio-temporal variation of the source direction. Processing of short blocks of data in reverberant environment can lead to clusters whose centroids correspond to spurious DOAs away from the source direction. We therefore propose in this paper a novel multi-detection source tracking approach that estimates the smoothed trajectory of the source DOAs. Results for realistic room simulations validate the proposed approach and demonstrate significant improvements in estimation accuracy compared to single-detection tracking.