English

Differentiable Tracking-Based Training of Deep Learning Sound Source Localizers

Audio and Speech Processing 2021-11-02 v1 Sound

Abstract

Data-based and learning-based sound source localization (SSL) has shown promising results in challenging conditions, and is commonly set as a classification or a regression problem. Regression-based approaches have certain advantages over classification-based, such as continuous direction-of-arrival estimation of static and moving sources. However, multi-source scenarios require multiple regressors without a clear training strategy up-to-date, that does not rely on auxiliary information such as simultaneous sound classification. We investigate end-to-end training of such methods with a technique recently proposed for video object detectors, adapted to the SSL setting. A differentiable network is constructed that can be plugged to the output of the localizer to solve the optimal assignment between predictions and references, optimizing directly the popular CLEAR-MOT tracking metrics. Results indicate large improvements over directly optimizing mean squared errors, in terms of localization error, detection metrics, and tracking capabilities.

Keywords

Cite

@article{arxiv.2111.00030,
  title  = {Differentiable Tracking-Based Training of Deep Learning Sound Source Localizers},
  author = {Sharath Adavanne and Archontis Politis and Tuomas Virtanen},
  journal= {arXiv preprint arXiv:2111.00030},
  year   = {2021}
}

Comments

Submitted to IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA2021)

R2 v1 2026-06-24T07:18:25.233Z