DNN and CNN with Weighted and Multi-task Loss Functions for Audio Event Detection
arXiv:1708.03211
Abstract
This report presents our audio event detection system submitted for Task 2, "Detection of rare sound events", of DCASE 2017 challenge. The proposed system is based on convolutional neural networks (CNNs) and deep neural networks (DNNs) coupled with novel weighted and multi-task loss functions and state-of-the-art phase-aware signal enhancement. The loss functions are tailored for audio event detection in audio streams. The weighted loss is designed to tackle the common issue of imbalanced data in background/foreground classification while the multi-task loss enables the networks to simultaneously model the class distribution and the temporal structures of the target events for recognition. Our proposed systems significantly outperform the challenge baseline, improving F-score from 72.7% to 90.0% and reducing detection error rate from 0.53 to 0.18 on average on the development data. On the evaluation data, our submission obtains an average F1-score of 88.3% and an error rate of 0.22 which are significantly better than those obtained by the DCASE baseline (i.e. an F1-score of 64.1% and an error rate of 0.64).
arXiv commentDCASE 2017 technical report
References in corpus (5)
- An Overview of Multi-Task Learning in Deep Neural Networks
- Recurrent Neural Networks for Polyphonic Sound Event Detection in Real Life Recordings
- Deep CNN Framework for Audio Event Recognition using Weakly Labeled Web Data
- CaR-FOREST: Joint Classification-Regression Decision Forests for Overlapping Audio Event Detection
- Learning Compact Structural Representations for Audio Events Using Regressor Banks
Cited by in corpus (5)
- A Combined Data-driven and Physics-driven Method for Steady Heat Conduction Prediction using Deep Convolutional Neural Networks
- Enabling Early Audio Event Detection with Neural Networks
- Semi-supervised learning for medical image classification using imbalanced training data
- Wrapped Loss Function for Regularizing Nonconforming Residual Distributions
- Adaptive Multi-scale Detection of Acoustic Events