Pushing the Decision Boundaries: Discovering New Classes in Audio Data
Keywords: Data-centric ML, Acoustic Signal Processing, Task Discovery
TL;DR: The paper studies the practicality of performing task discovery to extract class boundaries beyond the annotated class labels.
Abstract: We usually need new data to train or fine-tune machine learning models for new tasks. However, previously collected data might include relevant information that is enough to learn the desired tasks. In this paper, we explore discovering new classes in audio data by extending a recent vision-based task discovery framework with an audio processing pipeline. Our proposed pipeline aims to find new class boundaries on specific acoustic components, such as speech and background noise, which extends the vision-based framework to effectively handle audio data. Furthermore, we introduce a new metric for assessing the clarity of newly discovered class boundaries.
We show that, compared to the baseline task discovery framework, we can discover new classes with 21% higher clarity, in average.
Primary Subject Area: Active learning, Data cleaning, acquisition for ML
Paper Type: Extended abstracts: up to 2 pages
Participation Mode: In-person
Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.
Submission Number: 40
Loading