Learning Multiple Sound Source 2D Localization
This work provides an incremental improvement in sound source localization for applications requiring precise spatial awareness in enclosed environments.
This paper addresses the problem of localizing multiple sound sources in 2D Cartesian coordinates using multiple microphone arrays. The authors propose a deep learning-based encoding-decoding architecture with two improvements and two novel localization representations, achieving improved accuracy over a previous baseline.
In this paper, we propose novel deep learning based algorithms for multiple sound source localization. Specifically, we aim to find the 2D Cartesian coordinates of multiple sound sources in an enclosed environment by using multiple microphone arrays. To this end, we use an encoding-decoding architecture and propose two improvements on it to accomplish the task. In addition, we also propose two novel localization representations which increase the accuracy. Lastly, new metrics are developed relying on resolution-based multiple source association which enables us to evaluate and compare different localization approaches. We tested our method on both synthetic and real world data. The results show that our method improves upon the previous baseline approach for this problem.