# Self-training with noisy student model and semi-supervised loss function for dcase 2021 challenge task 4

arXiv: https://arxiv.org/abs/2107.02569 · HTML: https://mlsanity.com/paper/2107.02569/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 6 Jul 2021.

## Builds on (most cited first, 9 of 9)

- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results](https://mlsanity.com/paper/1703.01780/) (2017)
- [Automated Audio Captioning with Recurrent Neural Networks](https://mlsanity.com/paper/1706.10006/) (2017)
- [A Framework for the Robust Evaluation of Sound Event Detection](https://mlsanity.com/paper/1910.08440/) (2019)
- [Sound Event Detection and Separation: a Benchmark on Desed Synthetic Soundscapes](https://mlsanity.com/paper/2011.00801/) (2020)
- [SONYC: A System for the Monitoring, Analysis and Mitigation of Urban Noise Pollution](https://mlsanity.com/paper/1805.00889/) (2018)

## Built on by (most cited first, 2 of 2)

- [RCT: Random Consistency Training for Semi-supervised Sound Event Detection](https://mlsanity.com/paper/2110.11144/) (2021)
- [A benchmark of state-of-the-art sound event detection systems evaluated on synthetic soundscapes](https://mlsanity.com/paper/2202.01487/) (2022)

## Abstract (arXiv)

This report proposes a polyphonic sound event detection (SED) method for the DCASE 2021 Challenge Task 4. The proposed SED model consists of two stages: a mean-teacher model for providing target labels regarding weakly labeled or unlabeled data and a self-training-based noisy student model for predicting strong labels for sound events. The mean-teacher model, which is based on the residual convolutional recurrent neural network (RCRNN) for the teacher and student model, is first trained using all the training data from a weakly labeled dataset, an unlabeled dataset, and a strongly labeled synthetic dataset. Then, the trained mean-teacher model predicts the strong label to each of the weakly labeled and unlabeled datasets, which is brought to the noisy student model in the second stage of the proposed SED model. Here, the structure of the noisy student model is identical to the RCRNN-based student model of the mean-teacher model in the first stage. Then, it is self-trained by adding feature noises, such as time-frequency shift, mixup, SpecAugment, and dropout-based model noise. In addition, a semi-supervised loss function is applied to train the noisy student model, which acts as label noise injection. The performance of the proposed SED model is evaluated on the validation set of the DCASE 2021 Challenge Task 4, and then, several ensemble models that combine five-fold validation models with different hyperparameters of the semi-supervised loss function are finally selected as our final models.