Multi-stream Network With Temporal Attention For Environmental Sound Classification
This work addresses robustness issues in environmental sound classification for applications like audio scene analysis, though it appears incremental as it builds on existing multi-stream and attention methods.
The paper tackles the problem of robust environmental sound classification across varying tasks and temporal structures by introducing a multi-stream convolutional neural network with temporal attention, achieving new state-of-the-art performance on three datasets without architectural changes.
Environmental sound classification systems often do not perform robustly across different sound classification tasks and audio signals of varying temporal structures. We introduce a multi-stream convolutional neural network with temporal attention that addresses these problems. The network relies on three input streams consisting of raw audio and spectral features and utilizes a temporal attention function computed from energy changes over time. Training and classification utilizes decision fusion and data augmentation techniques that incorporate uncertainty. We evaluate this network on three commonly used data sets for environmental sound and audio scene classification and achieve new state-of-the-art performance without any changes in network architecture or front-end preprocessing, thus demonstrating better generalizability.