Few-shot bioacoustic event detection at the DCASE 2023 challenge

06/15/2023
by   Inês Nolasco, et al.
0

Few-shot bioacoustic event detection consists in detecting sound events of specified types, in varying soundscapes, while having access to only a few examples of the class of interest. This task ran as part of the DCASE challenge for the third time this year with an evaluation set expanded to include new animal species, and a new rule: ensemble models were no longer allowed. The 2023 few shot task received submissions from 6 different teams with F-scores reaching as high as 63 focusing on describing the elements that differed from previous years. We also take a look back at past editions to describe how the task has evolved. Not only have the F-score results steadily improved (40 type of systems proposed have also become more complex. Sound event detection systems are no longer simple variations of the baselines provided: multiple few-shot learning methodologies are still strong contenders for the task.

READ FULL TEXT
research
07/14/2022

Few-shot bioacoustic event detection at the DCASE 2022 challenge

Few-shot sound event detection is the task of detecting sound events, de...
research
07/28/2021

Proposal-based Few-shot Sound Event Detection for Speech and Environmental Sounds with Perceivers

There are many important applications for detecting and localizing speci...
research
09/16/2023

Regularized Contrastive Pre-training for Few-shot Bioacoustic Sound Detection

Bioacoustic sound event detection allows for better understanding of ani...
research
05/22/2023

Learning to detect an animal sound from five examples

Automatic detection and classification of animal sounds has many applica...
research
07/21/2022

Surrey System for DCASE 2022 Task 5: Few-shot Bioacoustic Event Detection with Segment-level Metric Learning

Few-shot audio event detection is a task that detects the occurrence tim...
research
10/05/2019

Few-shot tweet detection in emerging disaster events

Social media sources can provide crucial information in crisis situation...
research
10/09/2021

A Mutual learning framework for Few-shot Sound Event Detection

Although prototypical network (ProtoNet) has proved to be an effective m...

Please sign up or login with your details

Forgot password? Click here to reset