An End-to-End Audio Classification System based on Raw Waveforms and Mix-Training Strategy
This work addresses the challenge of overlapping sound events in audio classification, which is incremental as it builds on existing ResNet and attention methods.
The paper tackles multi-label audio classification by proposing an end-to-end system using raw waveforms and a mix-training strategy, achieving a mean average precision of 37.2% on the Audio Set dataset and outperforming the state-of-the-art multi-level attention model without extra training data.
Audio classification can distinguish different kinds of sounds, which is helpful for intelligent applications in daily life. However, it remains a challenging task since the sound events in an audio clip is probably multiple, even overlapping. This paper introduces an end-to-end audio classification system based on raw waveforms and mix-training strategy. Compared to human-designed features which have been widely used in existing research, raw waveforms contain more complete information and are more appropriate for multi-label classification. Taking raw waveforms as input, our network consists of two variants of ResNet structure which can learn a discriminative representation. To explore the information in intermediate layers, a multi-level prediction with attention structure is applied in our model. Furthermore, we design a mix-training strategy to break the performance limitation caused by the amount of training data. Experiments show that the mean average precision of the proposed audio classification system on Audio Set dataset is 37.2%. Without using extra training data, our system exceeds the state-of-the-art multi-level attention model.