Comparing Time and Frequency Domain for Audio Event Recognition Using Deep Learning
This addresses the feature representation problem for audio event recognition systems, though it appears incremental as it compares established domains rather than introducing new paradigms.
The paper investigated whether deep neural networks learn more discriminative features from time-domain or frequency-domain audio representations for acoustic event recognition. Results showed frequency-domain features are superior, and adding convolution/pooling layers achieved state-of-the-art performance on Freiburg-106 and ESC-10 datasets.
Recognizing acoustic events is an intricate problem for a machine and an emerging field of research. Deep neural networks achieve convincing results and are currently the state-of-the-art approach for many tasks. One advantage is their implicit feature learning, opposite to an explicit feature extraction of the input signal. In this work, we analyzed whether more discriminative features can be learned from either the time-domain or the frequency-domain representation of the audio signal. For this purpose, we trained multiple deep networks with different architectures on the Freiburg-106 and ESC-10 datasets. Our results show that feature learning from the frequency domain is superior to the time domain. Moreover, additionally using convolution and pooling layers, to explore local structures of the audio signal, significantly improves the recognition performance and achieves state-of-the-art results.