Learning from Between-class Examples for Deep Sound Recognition

Abstract

Deep learning methods have achieved high performance in sound recognition tasks. Deciding how to feed the training data is important for further performance improvement. We propose a novel learning method for deep sound recognition: Between-Class learning (BC learning). Our strategy is to learn a discriminative feature space by recognizing the between-class sounds as between-class sounds. We generate between-class sounds by mixing two sounds belonging to different classes with a random ratio. We then input the mixed sound to the model and train the model to output the mixing ratio. The advantages of BC learning are not limited only to the increase in variation of the training data; BC learning leads to an enlargement of Fisher's criterion in the feature space and a regularization of the positional relationship among the feature distributions of the classes. The experimental results show that BC learning improves the performance on various sound recognition networks, datasets, and data augmentation schemes, in which BC learning proves to be always beneficial. Furthermore, we construct a new deep sound recognition network (EnvNet-v2) and train it with BC learning. As a result, we achieved a performance surpasses the human level.

13 pages, 6 figures, published as a conference paper at ICLR 2018

References in corpus (2)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• SoundNet: Learning Sound Representations from Unlabeled Video

Cited by in corpus (26)

• Deep Learning for Audio Signal Processing
• Data Augmentation by Pairing Samples for Images Classification
• PSLA: Improving Audio Tagging with Pretraining, Sampling, Labeling, and Aggregation
• An Ensemble of Convolutional Neural Networks for Audio Classification
• FMix: Enhancing Mixed Sample Data Augmentation
• An overview of mixing augmentation methods and augmentation strategies
• Environmental Sound Classification on the Edge: A Pipeline for Deep Acoustic Networks on Extremely Resource-Constrained Devices
• Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attacks
• AST: Audio Spectrogram Transformer
• Universal Adversarial Audio Perturbations
• Pruning vs XNOR-Net: A Comprehensive Study of Deep Learning for Audio Classification on Edge-devices
• ASiT: Local-Global Audio Spectrogram vIsion Transformer for Event Classification
• CLAR: Contrastive Learning of Auditory Representations
• Attention based Convolutional Recurrent Neural Network for Environmental Sound Classification
• Inception-Based Network and Multi-Spectrogram Ensemble Applied For Predicting Respiratory Anomalies and Lung Diseases
• Data Interpolating Prediction: Alternative Interpretation of Mixup
• Between-class Learning for Image Classification
• Study of positional encoding approaches for Audio Spectrogram Transformers
• Deep Convolutional Neural Network with Mixup for Environmental Sound Classification
• Neural Networks Are More Productive Teachers Than Human Raters: Active Mixup for Data-Efficient Knowledge Distillation from a Blackbox Model
• A Unified Mixture-View Framework for Unsupervised Representation Learning
• Multi-class Novelty Detection Using Mix-up Technique
• RCT: Random Consistency Training for Semi-supervised Sound Event Detection
• Deep Epidemiological Modeling by Black-box Knowledge Distillation: An Accurate Deep Learning Model for COVID-19
• Learning Joint Embedding for Cross-Modal Retrieval
• Sub-Spectrogram Segmentation for Environmental Sound Classification via Convolutional Recurrent Neural Network and Score Level Fusion