CNN architectures for large-scale audio classification (2016-09-29T00:00:00.000000Z)

TL;DR

This work uses various CNN architectures to classify the soundtracks of a dataset of 70M training videos with 30,871 video-level labels, and investigates varying the size of both training set and label vocabulary, finding that analogs of the CNNs used in image classification do well on the authors' audio classification task, and larger training and label sets help up to a point.

Abstract

Convolutional Neural Networks (CNNs) have proven very effective in image classification and show promise for audio. We use various CNN architectures to classify the soundtracks of a dataset of 70M training videos (5.24 million hours) with 30,871 video-level labels. We examine fully connected Deep Neural Networks (DNNs), AlexNet [1], VGG [2], Inception [3], and ResNet [4]. We investigate varying the size of both training set and label vocabulary, finding that analogs of the CNNs used in image classification do well on our audio classification task, and larger training and label sets help up to a point. A model using embeddings from these classifiers does much better than raw features on the Audio Set [5] Acoustic Event Detection (AED) classification task.

Authors

Shawn Hershey

1 papers

Sourish Chaudhuri

3 papers

D. Ellis

6 papers

TL;DR

Abstract

Authors

References31 items

What's In A Name? Malay Seals As Onomastic Sources

Audio Set: An ontology and human-labeled dataset for audio events

TRECVID 2016: Evaluating Video Search, Video Event Detection, Localization, and Hyperlinking

TUT database for acoustic scene classification and sound event detection

Audio Event Detection using Weakly Labeled Data

Deep Convolutional Neural Networks and Data Augmentation for Acoustic Event Detection

Recurrent neural networks for polyphonic sound event detection in real life recordings

Deep Residual Learning for Image Recognition

Rethinking the Inception Architecture for Computer Vision

Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks

ActivityNet: A large-scale video benchmark for human activity understanding

Beyond short snippets: Deep networks for video classification

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Very Deep Convolutional Networks for Large-Scale Image Recognition

Large-Scale Video Classification with Convolutional Neural Networks

ImageNet classification with deep convolutional neural networks

Machine Hearing: An Emerging Field [Exploratory DSP]

Real-world acoustic event detection

Acoustic event detection in real life recordings

Rectified Linear Units Improve Restricted Boltzmann Machines

ImageNet: A large-scale hierarchical image database

CLEAR Evaluation of Acoustic Event Detection and Classification Systems

Retrieval evaluation with incomplete information

CP-JKU SUBMISSIONS FOR DCASE-2016 : A HYBRID APPROACH USING BINAURAL I-VECTORS AND DEEP CONVOLUTIONAL NEURAL NETWORKS

TensorFlow: Large-scale machine learning on heterogeneous systems

Dropout: a simple way to prevent neural networks from overfitting

AN EXEMPLAR-BASED NMF APPROACH FOR AUDIO EVENT DETECTION

Introducing the knowledge graph: things, not strings

Machine Hearing: An Emerging Field

ROC Graphs: Notes and Practical Considerations for Researchers

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names