SELD-TCN: Sound Event Localization & Detection via Temporal Convolutional Networks
This work addresses the need for efficient acoustic perception in autonomous robotic systems, such as self-driving cars, by providing a hardware-friendly solution, though it is incremental as it builds on existing SELD frameworks.
The paper tackled the problem of sound event localization and detection (SELD) for autonomous systems by proposing a temporal convolutional network (TCN) architecture, which outperformed the state-of-the-art SELDnet on four datasets and achieved 4x faster training and 40x faster inference times on a GPU.
The understanding of the surrounding environment plays a critical role in autonomous robotic systems, such as self-driving cars. Extensive research has been carried out concerning visual perception. Yet, to obtain a more complete perception of the environment, autonomous systems of the future should also take acoustic information into account. Recent sound event localization and detection (SELD) frameworks utilize convolutional recurrent neural networks (CRNNs). However, considering the recurrent nature of CRNNs, it becomes challenging to implement them efficiently on embedded hardware. Not only are their computations strenuous to parallelize, but they also require high memory bandwidth and large memory buffers. In this work, we develop a more robust and hardware-friendly novel architecture based on a temporal convolutional network(TCN). The proposed framework (SELD-TCN) outperforms the state-of-the-art SELDnet performance on four different datasets. Moreover, SELD-TCN achieves 4x faster training time per epoch and 40x faster inference time on an ordinary graphics processing unit (GPU).