Multi-ACCDOA: Localizing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Training
Authors:
Preprint
Abstract
Sound event localization and detection (SELD) involves identifying the direction-of-arrival (DOA) and the event class. The SELD methods with a class-wise output format make the model predict activities of all sound event classes and corresponding locations. The class-wise methods can output activity-coupled Cartesian DOA (ACCDOA) vectors, which enable us to solve a SELD task with a single target using a single network. However, there is still a challenge in detecting the same event class from multiple locations. To overcome this problem while maintaining the advantages of the class-wise format, we extended ACCDOA to a multi one and proposed auxiliary duplicating permutation invariant training (ADPIT). The multi- ACCDOA format (a class- and track-wise output format) enables the model to solve the cases with overlaps from the same class. The class-wise ADPIT scheme enables each track of the multi-ACCDOA format to learn with the same target as the single-ACCDOA format. In evaluations with the DCASE 2021 Task 3 dataset, the model trained with the multi-ACCDOA format and with the class-wise ADPIT detects overlapping events from the same class while maintaining its performance in the other cases. Also, the proposed method performed comparably to state-of-the-art SELD methods with fewer parameters.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- A Four-Stage Data Augmentation Approach to ResNet-Conformer Based Acoustic Modeling for Sound Event Localization and DetectionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
- Overview and Evaluation of Sound Event Localization and Detection in DCASE 2019IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020
- Sound Event Localization and Detection of Overlapping Sources Using Convolutional Recurrent Neural NetworksIEEE Journal of Selected Topics in Signal Processing, 2018
- Investigations on Data Augmentation and Loss Functions for Deep Learning Based Speech-Background Separation2018
- AENet: Learning Deep Audio Features for Video AnalysisIEEE Transactions on Multimedia, 2017
- Multitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural NetworksIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2017
- Sound Source Localization Using Deep Learning ModelsJournal of Robotics and Mechatronics, 2017
- Environmental Sound Recognition With Time–Frequency Audio FeaturesIEEE Transactions on Audio, Speech, and Language Processing, 2009