Spoken content, such as online videos and podcasts, often spans multiple topics, which makes automatic topic segmentation essential for user navigation and downstream applications. However, current methods do not fully leverage acoustic features, leaving room for improvement. We propose a multi-modal approach that fine-tunes both a text encoder and a Siamese audio encoder, capturing acoustic cues around sentence boundaries. Experiments on a large-scale dataset of YouTube videos show substantial gains over text-only and multi-modal baselines. Our model also proves more resilient to ASR noise and outperforms a larger text-only baseline on three additional datasets in Portuguese, German, and English, underscoring the value of learned acoustic features for robust topic segmentation.
@article{arxiv.2602.06647,
title = {Reading Between the Waves: Robust Topic Segmentation Using Inter-Sentence Audio Features},
author = {Steffen Freisinger and Philipp Seeberger and Tobias Bocklet and Korbinian Riedhammer},
journal= {arXiv preprint arXiv:2602.06647},
year = {2026}
}