Robust Speech Activity Detection in the Presence of Singing Voice
Abstract
Speech Activity Detection (SAD) systems often misclassify singing as speech, leading to degraded performance in applications such as dialogue enhancement and automatic speech recognition. We introduce Singing-Robust Speech Activity Detection ( SR-SAD ), a neural network designed to robustly detect speech in the presence of singing. Our key contributions are: i) a training strategy using controlled ratios of speech and singing samples to improve discrimination, ii) a computationally efficient model that maintains robust performance while reducing inference runtime, and iii) a new evaluation metric tailored to assess SAD robustness in mixed speech-singing scenarios. Experiments on a challenging dataset spanning multiple musical genres show that SR-SAD maintains high speech detection accuracy (AUC = 0.919) while rejecting singing. By explicitly learning to distinguish between speech and singing, SR-SAD enables more reliable SAD in mixed speech-singing scenarios.
Cite
@article{arxiv.2512.09713,
title = {Robust Speech Activity Detection in the Presence of Singing Voice},
author = {Philipp Grundhuber and Mhd Modar Halimeh and Martin Strauß and Emanuël A. P. Habets},
journal= {arXiv preprint arXiv:2512.09713},
year = {2025}
}
Comments
This paper has been published in: 2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)