2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

Paper Detail

Paper IDAUD-12.6
Paper Title PROTOTYPICAL NETWORKS FOR DOMAIN ADAPTATION IN ACOUSTIC SCENE CLASSIFICATION
Authors Shubhr Singh, Helen L. Bear, Emmanouil Benetos, Queen Mary University of London, United Kingdom
SessionAUD-12: Detection and Classification of Acoustic Scenes and Events 1: Few-shot learning
LocationGather.Town
Session Time:Wednesday, 09 June, 15:30 - 16:15
Presentation Time:Wednesday, 09 June, 15:30 - 16:15
Presentation Poster
Topic Audio and Acoustic Signal Processing: [AUD-CLAS] Detection and Classification of Acoustic Scenes and Events
IEEE Xplore Open Preview  Click here to view in IEEE Xplore
Virtual Presentation  Click here to watch in the Virtual Conference
Abstract Acoustic Scene Classification (ASC) refers to the task of assigning a semantic label to an audio stream that characterises the environment in which it was recorded. In recent times, Deep Neural Networks(DNNs) have emerged as the model of choice for ASC. However, in real world scenarios, domain adaptation remains a persistent problem for ASC models. In the search for an optimal solution to the said problem, we explore a metric learning approach called prototypical networks using the TUT Urban Acoustic Scenes dataset, which consists of 10 different acoustic scenes recorded across 10 cities. In order to replicate the domain adaptation scenario, we divide the dataset into source domain data consisting of data samples from eight randomly selected cities and target domain data consisting of data from the remaining two cities. We evaluate the performance of the net-work against a selected baseline network under various experimental scenarios and based on the results we conclude that metric learning is a promising approach towards addressing the domain adaptation problem in ASC.