Sound source detection, localization and classification using consecutive ensemble of CRNN models

Abstract

In this paper, we describe our method for DCASE2019 task3: Sound Event Localization and Detection (SELD). We use four CRNN SELDnet-like single output models which run in a consecutive manner to recover all possible information of occurring events. We decompose the SELD task into estimating number of active sources, estimating direction of arrival of a single source, estimating direction of arrival of the second source where the direction of the first one is known and a multi-label classification task. We use custom consecutive ensemble to predict events' onset, offset, direction of arrival and class. The proposed approach is evaluated on the TAU Spatial Sound Events 2019 - Ambisonic and it is compared with other participants' submissions.

5 pages, 3 figures, conference

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
• Squeeze-and-Excitation Networks
• Sound Event Localization and Detection of Overlapping Sources Using Convolutional Recurrent Neural Networks
• Polyphonic Sound Event Detection and Localization using a Two-Stage Strategy
• A multi-room reverberant dataset for sound event localization and detection

Cited by in corpus (4)

• Audiovisual Transformer Architectures for Large-Scale Classification and Synchronization of Weakly Labeled Audio Events
• Joint framework with deep feature distillation and adaptive focal loss for weakly supervised audio tagging and acoustic event detection
• WhisperMask: A Noise Suppressive Mask-Type Microphone for Whisper Speech
• Capturing scattered discriminative information using a deep architecture in acoustic scene classification