Acoustic Event Detection from Weakly Labeled Data Using Auditory Salience

Podwinska, Z; Sobieraj, I; Fazenda, BM; Davies, WJ; Plumbley, MD

Abstract

Acoustic Event Detection (AED) is an important task of machine listening which, in recent years, has been addressed using common machine learning methods like Non-negative Matrix Factorization (NMF) or deep learning. However, most of these approaches do not take into consideration the way that human auditory system detects
salient sounds. In this work, we propose a method for AED using weakly labeled data that combines a Non-negative Matrix Factorization model with a salience model based on predictive coding in the form of Kalman filters. We show that models of auditory perception, particularly auditory salience, can be successfully incorporated into existing AED methods and improve their performance on rare event
detection. We evaluate the method on the Task2 of DCASE2017 Challenge.

Citation

Podwinska, Z., Sobieraj, I., Fazenda, B., Davies, W., & Plumbley, M. (2019, May). Acoustic Event Detection from Weakly Labeled Data Using Auditory Salience. Presented at 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK

• Presentation Conference Type: Other
• Conference Name: 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• Conference Location: Brighton, UK
• Start Date: May 12, 2019
• End Date: May 17, 2019
• Online Publication Date: Apr 17, 2019
• Publication Date: Apr 17, 2019
• Deposit Date: May 7, 2019
• Publicly Available Date: May 7, 2019
• Book Title: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• ISBN: 9781479981311
• DOI: https://doi.org/10.1109/ICASSP.2019.8683586
• Publisher URL: https://doi.org/10.1109/ICASSP.2019.8683586
• Additional Information: Event Type : Conference; Funders : Engineering and Physical Sciences Research Council (EPSRC);European Union; Projects : Acoustic event detection from weakly labeled data using auditory salience; Grant Number: EP/N014111/1; Grant Number: H2020-MSCA-ITN-2014 642685