Polyphonic sound event detection based on convolutional recurrent neural networks with semi-supervised loss function for DCASE challenge 2020 task 4
This work addresses sound event detection for audio analysis applications, but it is incremental as it builds on existing baseline models and focuses on a specific challenge task.
The paper tackles polyphonic sound event detection for the DCASE 2020 Challenge Task 4 by proposing a method based on semi-supervised learning to handle weakly labeled, unlabeled, and synthetic datasets, achieving results through an ensemble model evaluated on a validation set.
This report proposes a polyphonic sound event detection (SED) method for the DCASE 2020 Challenge Task 4. The proposed SED method is based on semi-supervised learning to deal with the different combination of training datasets such as weakly labeled dataset, unlabeled dataset, and strongly labeled synthetic dataset. Especially, the target label of each audio clip from weakly labeled or unlabeled dataset is first predicted by using the mean teacher model that is the DCASE 2020 baseline. The data with predicted labels are used for training the proposed SED model, which consists of CNNs with skip connections and self-attention mechanism, followed by RNNs. In order to compensate for the erroneous prediction of weakly labeled and unlabeled data, a semi-supervised loss function is employed for the proposed SED model. In this work, several versions of the proposed SED model are implemented and evaluated on the validation set according to the different parameter setting for the semi-supervised loss function, and then an ensemble model that combines five-fold validation models is finally selected as our final model.