Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sound event localization and detection based on crnn using rectangular filters and channel rotation data augmentation

Francesca Ronchini, Daniel Arteaga, Andrés Pérez-López

TL;DR

The proposed system is a convolutional recurrent neural network using rectangular filters specialized in recognizing significant spectral features related to the task, considerably improving Error Rate and F-score for location-aware detection.

Abstract

Sound Event Localization and Detection refers to the problem of identifying the presence of independent or temporally-overlapped sound sources, correctly identifying to which sound class it belongs, estimating their spatial directions while they are active. In the last years, neural networks have become the prevailing method for sound Event Localization and Detection task, with convolutional recurrent neural networks being among the most used systems. This paper presents a system submitted to the Detection and Classification of Acoustic Scenes and Events 2020 Challenge Task 3. The algorithm consists of a convolutional recurrent neural network using rectangular filters, specialized in recognizing significant spectral features related to the task. In order to further improve the score and to generalize the system performance to unseen data, the training dataset size has been increased using data augmentation. The technique used for that is based on channel rotations and reflection on the xy plane in the First Order Ambisonic domain, which allows improving Direction of Arrival labels keeping the physical relationships between channels. Evaluation results on the development dataset show that the proposed system outperforms the baseline results, considerably improving Error Rate and F-score for location-aware detection.

Sound event localization and detection based on crnn using rectangular filters and channel rotation data augmentation

TL;DR

The proposed system is a convolutional recurrent neural network using rectangular filters specialized in recognizing significant spectral features related to the task, considerably improving Error Rate and F-score for location-aware detection.

Abstract

Sound Event Localization and Detection refers to the problem of identifying the presence of independent or temporally-overlapped sound sources, correctly identifying to which sound class it belongs, estimating their spatial directions while they are active. In the last years, neural networks have become the prevailing method for sound Event Localization and Detection task, with convolutional recurrent neural networks being among the most used systems. This paper presents a system submitted to the Detection and Classification of Acoustic Scenes and Events 2020 Challenge Task 3. The algorithm consists of a convolutional recurrent neural network using rectangular filters, specialized in recognizing significant spectral features related to the task. In order to further improve the score and to generalize the system performance to unseen data, the training dataset size has been increased using data augmentation. The technique used for that is based on channel rotations and reflection on the xy plane in the First Order Ambisonic domain, which allows improving Direction of Arrival labels keeping the physical relationships between channels. Evaluation results on the development dataset show that the proposed system outperforms the baseline results, considerably improving Error Rate and F-score for location-aware detection.

Paper Structure

This paper contains 9 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. METHOD
  3. Feature Extraction
  4. The network
  5. Data augmentation
  6. Hyper-parameters
  7. Experiments
  8. Results and discussion
  9. Conclusions and Future Work

Figures (2)

  • Figure 1: The proposed network architecture.
  • Figure 2: Example of augmented intensity vector. Pattern applied: reflection with respect to xy plane. The image shows the Z-component of the intensity vector, before and after applying a reflection on the Z-channel. The horizontal axis represents time (expressed in frames) and the vertical axis mel-band coefficients.