Secost: Sequential co-supervision for large scale weakly labeled audio event detection
This work addresses the challenge of robust audio event detection with minimal annotation for large-scale applications, representing an incremental improvement over existing methods.
The paper tackles the problem of scaling audio event detection to hundreds of sound categories using weakly supervised learning, proposing SeCoST, a framework that achieves a mean average precision of 0.383 on Audioset, outperforming prior state-of-the-art methods.
Weakly supervised learning algorithms are critical for scaling audio event detection to several hundreds of sound categories. Such learning models should not only disambiguate sound events efficiently with minimal class-specific annotation but also be robust to label noise, which is more apparent with weak labels instead of strong annotations. In this work, we propose a new framework for designing learning models with weak supervision by bridging ideas from sequential learning and knowledge distillation. We refer to the proposed methodology as SeCoST (pronounced Sequest) -- Sequential Co-supervision for training generations of Students. SeCoST incrementally builds a cascade of student-teacher pairs via a novel knowledge transfer method. Our evaluations on Audioset (the largest weakly labeled dataset available) show that SeCoST achieves a mean average precision of 0.383 while outperforming prior state of the art by a considerable margin.