DOA Estimation with Lightweight Network on LLM-Aided Simulated Acoustic Scenes
This work addresses DOA estimation for spatial audio and acoustic signal processing applications, but it is incremental as it builds on existing neural-based methods with a new dataset and model optimization.
The paper tackled the problem of Direction-of-Arrival (DOA) estimation by using a dataset synthesized with LLM assistance for more realistic and diverse acoustic scenes, and proposed LightDOA, a lightweight model that achieves satisfactory accuracy and robustness while maintaining low computational complexity.
Direction-of-Arrival (DOA) estimation is critical in spatial audio and acoustic signal processing, with wide-ranging applications in real-world. Most existing DOA models are trained on synthetic data by convolving clean speech with room impulse responses (RIRs), which limits their generalizability due to constrained acoustic diversity. In this paper, we revisit DOA estimation using a recently introduced dataset constructed with the assistance of large language models (LLMs), which provides more realistic and diverse spatial audio scenes. We benchmark several representative neural-based DOA methods on this dataset and propose LightDOA, a lightweight DOA estimation model based on depthwise separable convolutions, specifically designed for mutil-channel input in varying environments. Experimental results show that LightDOA achieves satisfactory accuracy and robustness across various acoustic scenes while maintaining low computational complexity. This study not only highlights the potential of spatial audio synthesized with the assistance of LLMs in advancing robust and efficient DOA estimation research, but also highlights LightDOA as efficient solution for resource-constrained applications.