Supervised Representation Learning for Audio Scene Classification
Author:
- 23 May 2017
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 25 (6) , 1253-1265
Abstract
This paper investigates the use of supervised feature learning approaches for extracting relevant and discriminative features from acoustic scene recordings. Owing to the recent release of open datasets for acoustic scene classification problems, representation learning techniques can now be envisioned for solving the problem of feature extraction. This paper makes a step toward this goal by first introducing a supervised nonnegative matrix factorization (SNMF). Our goal through this SNMF is to induce the matrix decomposition to carry out discriminative information in addition to the usual generative ones. We achieve this objective by augmenting the nonnegative matrix factorization optimization problem with a novel loss function related to class labels of each column of the matrix to decompose. While the scale of the datasets available is still small compared to those available in computer vision, we have studied models based on convolutional neural networks. We have analyzed the performances of these models on the DCASE-16 dataset and a corrected version of the LITIS Rouen one. Our experiments show that despite the small-scale setting, supervised feature learning is favorably competitive compared to the current state-of-the-art features. We also point out that for smaller scale dataset, SNMF is indeed slightly less prone to overfitting than convolutional neural networks. While the performances of these learned features are interesting per se, a deeper analysis of their behavior in the acoustic scene problem context raises open and difficult questions that we believe, need to be addressed for further performance breakthroughs.Keywords
Funding Information
- ANR Deep in France Project
- Regional Project GRR
- FEDER
- DAISI
This publication has 37 references indexed in Scilit:
- Histogram of gradients of Time-Frequency Representations for Audio Scene DetectionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Autoencoder-based Unsupervised Domain Adaptation for Speech Emotion RecognitionIEEE Signal Processing Letters, 2014
- Algorithms for nonnegative matrix and tensor factorizations: a unified view based on block coordinate descent frameworkJournal of Global Optimization, 2013
- Noise-Robust environmental sound classification method based on combination of ICA and MP featuresArtificial Intelligence Research, 2012
- Algorithms for Nonnegative Matrix Factorization with the β-DivergenceNeural Computation, 2011
- Fast Local Algorithms for Large Scale Nonnegative Matrix and Tensor FactorizationsIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 2009
- Supervised and Semi-supervised Separation of Sounds from Single-Channel Mixtures2007
- Projected Gradient Methods for Nonnegative Matrix FactorizationNeural Computation, 2007
- Acoustic environment classificationACM Transactions on Speech and Language Processing, 2006
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998