To separate speech signals using blind techniques, the DUET algorithm is used in which each source signal is separated by masking the mixed signals in the Time-Frequency domain. To do so, a two dimensional Histogram of mixed parameters is generated which is computationa More
To separate speech signals using blind techniques, the DUET algorithm is used in which each source signal is separated by masking the mixed signals in the Time-Frequency domain. To do so, a two dimensional Histogram of mixed parameters is generated which is computationally burden, and thus, can not be used in real-time. In this paper, we introduce a new algorithm in which the separation process can be carried out online. Also, simulation results show that this algorithm has a comparable precision with respect to the DUET algorithm.
Manuscript profile
Estimation of speakers' directions is one of the interesting topics. In this paper, we introduce the TFCS algorithm for direction-of-arrival estimation of multiple speakers. In this algorithm, no limitations are imposed on the environment and speech signals. Simulation More
Estimation of speakers' directions is one of the interesting topics. In this paper, we introduce the TFCS algorithm for direction-of-arrival estimation of multiple speakers. In this algorithm, no limitations are imposed on the environment and speech signals. Simulation results show that the proposed algorithm outperforms the recently addressed onset and spectral-based methods.
Manuscript profile
In this paper, we propose a new approach for selection of subsets of active sensors with some constraints on energy consumption and estimation error for tracking of a target. The proposed approach exploits the decentralized estimation by using the extended information f More
In this paper, we propose a new approach for selection of subsets of active sensors with some constraints on energy consumption and estimation error for tracking of a target. The proposed approach exploits the decentralized estimation by using the extended information filter for target tracking. Furthermore, a cost function is defined using spatial correlation for sensor selection. Consequently, the Spatial Split algorithm is proposed based on spatial correlation coefficients for sensor selection. At last, for high speed targets, we propose a modification on spatial split algorithm by changing the sensing range with respect to the target speed. Simulation results show that the tracking accuracy is analogous to those of optimal estimation methods. It is also found that energy consumption decreases due to activating only necessary sensors.
Manuscript profile
