Audio Set classification with attention model: A probabilistic perspective

Abstract

This paper investigates the classification of the Audio Set dataset. Audio Set is a large scale weakly labelled dataset of sound clips. Previous work used multiple instance learning (MIL) to classify weakly labelled data. In MIL, a bag consists of several instances, and a bag is labelled positive if at least one instances in the audio clip is positive. A bag is labelled negative if all the instances in the bag are negative. We propose an attention model to tackle the MIL problem and explain this attention model from a novel probabilistic perspective. We define a probability space on each bag, where each instance in the bag has a trainable probability measure for each class. Then the classification of a bag is the expectation of the classification output of the instances in the bag with respect to the learned probability measure. Experimental results show that our proposed attention model modeled by fully connected deep neural network obtains mAP of 0.327 on Audio Set dataset, outperforming the Google's baseline of 0.314 and recurrent neural network of 0.325.

Accepted by ICASSP 2018

References in corpus (4)

• Automatic tagging using deep convolutional neural networks
• Audio Event Detection using Weakly Labeled Data
• Unsupervised Feature Learning Based on Deep Models for Environmental Audio Tagging
• Large-scale weakly supervised audio classification using gated convolutional neural network

Cited by in corpus (5)

• Audio-Based Activities of Daily Living (ADL) Recognition with Large-Scale Acoustic Embeddings from Online Videos
• Multi-level Attention Model for Weakly Supervised Audio Classification
• An End-to-End Audio Classification System based on Raw Waveforms and Mix-Training Strategy
• SAM-GCNN: A Gated Convolutional Neural Network with Segment-Level Attention Mechanism for Home Activity Monitoring
• Multi-label Sound Event Retrieval Using a Deep Learning-based Siamese Structure with a Pairwise Presence Matrix