What Did I Just Hear? Detecting Pornographic Sounds in Adult Videos Using Neural Networks

TL;DR

This paper presents a neural network-based method for detecting pornographic sounds in videos, using spectral features and a voting technique to improve audio-level classification accuracy.

Contribution

It introduces a CNN trained on log mel spectrograms for pornographic sound detection and a voting segment-to-audio method for whole audio classification.

Findings

CNN on log mel spectrogram achieves top performance

Log mel spectrogram provides better feature representations

Voting segment-to-audio improves detection accuracy

Abstract

Audio-based pornographic detection enables efficient adult content filtering without sacrificing performance by exploiting distinct spectral characteristics. To improve it, we explore pornographic sound modeling based on different neural architectures and acoustic features. We find that CNN trained on log mel spectrogram achieves the best performance on Pornography-800 dataset. Our experiment results also show that log mel spectrogram allows better representations for the models to recognize pornographic sounds. Finally, to classify whole audio waveforms rather than segments, we employ voting segment-to-audio technique that yields the best audio-level detection results.