Unsupervised Feature Learning Based on Deep Models for Environmental Audio Tagging

Abstract

Environmental audio tagging aims to predict only the presence or absence of certain acoustic events in the interested acoustic scene. In this paper we make contributions to audio tagging in two parts, respectively, acoustic modeling and feature learning. We propose to use a shrinking deep neural network (DNN) framework incorporating unsupervised feature learning to handle the multi-label classification task. For the acoustic modeling, a large set of contextual frames of the chunk are fed into the DNN to perform a multi-label classification for the expected tags, considering that only chunk (or utterance) level rather than frame-level labels are available. Dropout and background noise aware training are also adopted to improve the generalization capability of the DNNs. For the unsupervised feature learning, we propose to use a symmetric or asymmetric deep de-noising auto-encoder (sDAE or aDAE) to generate new data-driven features from the Mel-Filter Banks (MFBs) features. The new features, which are smoothed against background noise and more compact with contextual information, can further improve the performance of the DNN baseline. Compared with the standard Gaussian Mixture Model (GMM) baseline of the DCASE 2016 audio tagging challenge, our proposed method obtains a significant equal error rate (EER) reduction from 0.21 to 0.13 on the development set. The proposed aDAE system can get a relative 6.7% EER reduction compared with the strong DNN baseline on the development set. Finally, the results also show that our approach obtains the state-of-the-art performance with 0.15 EER on the evaluation set of the DCASE 2016 audio tagging task while EER of the first prize of this challenge is 0.17.

10 pages, dcase 2016 challenge

References in corpus (6)

• Improving neural networks by preventing co-adaptation of feature detectors
• YFCC100M: The New Data in Multimedia Research
• YouTube-8M: A Large-Scale Video Classification Benchmark
• Automatic large-scale classification of bird sounds is strongly improved by unsupervised feature learning
• Automatic tagging using deep convolutional neural networks
• Audio Event Detection using Weakly Labeled Data

Cited by in corpus (15)

• WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Research
• Automated Audio Captioning: An Overview of Recent Progress and New Challenges
• General audio tagging with ensembling convolutional neural network and statistical features
• Sample Mixed-Based Data Augmentation for Domestic Audio Tagging
• Audio Tagging by Cross Filtering Noisy Labels
• ASiT: Local-Global Audio Spectrogram vIsion Transformer for Event Classification
• Spoken Language Identification using ConvNets
• Learning audio representations via phase prediction
• Audio Set classification with attention model: A probabilistic perspective
• Attention and Localization based on a Deep Convolutional Recurrent Model for Weakly Supervised Audio Tagging
• T-vectors: Weakly Supervised Speaker Identification Using Hierarchical Transformer Model
• Unsupervised Discriminative Learning of Sounds for Audio Event Classification
• Deep Learning for Audio Transcription on Low-Resource Datasets
• Enhancing Sound Texture in CNN-Based Acoustic Scene Classification
• Weakly Supervised Training of Hierarchical Attention Networks for Speaker Identification