{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multi-iteration-multi-stage-fine-tuning-of","title":"Multi-Iteration Multi-Stage Fine-Tuning of Transformers for Sound Event Detection with Heterogeneous Datasets","arxiv_id":"2407.12997","date":"2024-07-17","proceeding":null,"authors":["Florian Schmid","Paul Primus","Tobias Morocutti","Jonathan Greif","Gerhard Widmer"],"abstract":"A central problem in building effective sound event detection systems is the lack of high-quality, strongly annotated sound event datasets. For this reason, Task 4 of the DCASE 2024 challenge proposes learning from two heterogeneous datasets, including audio clips labeled with varying annotation granularity and with different sets of possible events. We propose a multi-iteration, multi-stage procedure for fine-tuning Audio Spectrogram Transformers on the joint DESED and MAESTRO Real datasets. The first stage closely matches the baseline system setup and trains a CRNN model while keeping the pre-trained transformer model frozen. In the second stage, both CRNN and transformer are fine-tuned using heavily weighted self-supervised losses. After the second stage, we compute strong pseudo-labels for all audio clips in the training set using an ensemble of fine-tuned transformers. Then, in a second iteration, we repeat the two-stage training process and include a distillation loss based on the pseudo-labels, achieving a new single-model, state-of-the-art performance on the public evaluation set of DESED with a PSDS1 of 0.692. A single model and an ensemble, both based on our proposed training procedure, ranked first in Task 4 of the DCASE Challenge 2024.","url_abs":"https://arxiv.org/abs/2407.12997v1","url_pdf":"https://arxiv.org/pdf/2407.12997v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multi-iteration-multi-stage-fine-tuning-of","repo_url":"https://github.com/cpjku/cpjku_dcase24","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"event-detection","task_name":"Event Detection"},{"task_slug":"sound-event-detection","task_name":"Sound Event Detection"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}