The Aalto system based on fine-tuned AudioSet features for DCASE2018 task2 - general purpose audio tagging

In this paper, we presented a neural network system for DCASE 2018 task 2, general purpose audio tagging. We fine-tuned the Google AudioSet feature generation model with different settings for the given 41 classes on top of a fully connected layer with 100 units. Then we used the fine-tuned models to generate 128 dimensional features for each 0.960s audio. We tried different neural network structures including LSTM and multi-level attention models. In our experiments, the multi-level attention model has shown its superiority over others. Truncating the silence parts, repeating and splitting the audio into the fixed length, pitch shifting augmentation, and mixup techniques are all used in our experiments. The proposed system achieved a result with MAP@3 score at 0.936, which outperforms the baseline result of 0.704 and achieves top 8% in the public leaderboard.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Hongyi Zhang,et al.  mixup: Beyond Empirical Risk Minimization , 2017, ICLR.

[3]  Biao Leng,et al.  A Multi-level Weighted Representation for Person Re-identification , 2017, ICANN.

[4]  Bin Yang,et al.  Multi-level attention model for weakly supervised audio classification , 2018, DCASE.

[5]  Daniel P. W. Ellis,et al.  General-purpose Tagging of Freesound Audio with AudioSet Labels: Task Description, Dataset, and Baseline , 2018, DCASE.

[6]  Yong Xu,et al.  Audio Set Classification with Attention Model: A Probabilistic Perspective , 2017, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[7]  Aren Jansen,et al.  CNN architectures for large-scale audio classification , 2016, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  Juhan Nam,et al.  Multi-Level and Multi-Scale Feature Aggregation Using Pretrained Convolutional Neural Networks for Music Auto-Tagging , 2017, IEEE Signal Processing Letters.

[9]  Hagai Aronowitz.  Segmental Modeling for Audio Segmentation , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[10]  Xavier Serra,et al.  Freesound Datasets: A Platform for the Creation of Open Audio Datasets , 2017, ISMIR.

[11]  Ankit Shah,et al.  DCASE2017 Challenge Setup: Tasks, Datasets and Baseline System , 2017, DCASE.

[12]  Aren Jansen,et al.  Audio Set: An ontology and human-labeled dataset for audio events , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).