Acoustic Scene Classification

This paper describes our contribution to the Acoustic Scene Classification task of the IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2017). We propose a system for this task using a recurrent sequence to sequence autoencoder for unsupervised representation learning from raw audio files. First, we extract mel-spectrograms from the raw audio files. Second, we train a recurrent sequence to sequence autoen-coder on these spectrograms, that are considered as time-dependent frequency vectors. Then, we extract, from a fully connected layer between the decoder and encoder units, the learnt representations of spectrograms as the feature vectors for the corresponding audio instances. Finally, we train a multilayer perceptron neural network on these feature vectors to predict the class labels. In comparison to the baseline, the accuracy is increased from 74.8 % to 88.0 % on the development set, and from 61.0 % to 67.5 % on the test set.

This paper describes our approach for DCASE 2019 Task2: Audio tagging with noisy labels and minimal supervision. This challenge uses a smaller set of manually labeled data and a larger set of noiselabeled data to enable the system to perform multi-label audio tagging tasks with minimal supervision conditions. We aim to tagging the audio clips with convolutional neural network under a limited computation and storage resources. To tackle the problem of noisy label data, we propose a data generation method named Dominate Mixup. It can restrain the impact of incorrect label during back propagation and it’s suitable for multi-class classification problem. In response to the variable length of audio clips, we conduct an efficient learning method with cyclical audio length which allow us to learn more pattern from widely diverse sound events. On the public leaderboard for the competition, our single model and simple ensemble of 5 models score 0.711 and 0.725 respectively.

A general problem in acoustic scene classification task is the mis-matched conditions between training and testing data, which significantly reduces the performance of the developed methods on classification accuracy. As a countermeasure, we present the first method of unsupervised adversarial domain adaptation for acoustic scene classification. We employ a model pre-trained on data from one set of conditions and by using data from other set of conditions, we adapt the model in order that its output cannot be used for classifying the set of conditions that input data belong to. We use a freely available dataset from the DCASE 2018 challenge Task 1, subtask B, that contains data from mismatched recording devices. We consider the scenario where the annotations are available for the data recorded from one device, but not for the rest. Our results show that with our model agnostic method we can achieve ∼ 10% increase at the accuracy on an unseen and unlabeled dataset, while keeping almost the same performance on the labeled dataset.