arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.06647cs.CLeess.AS
收藏

在波涛间阅读:利用跨句音频特征实现鲁棒性话题分割

Reading Between the Waves: Robust Topic Segmentation Using Inter-Sentence Audio Features

Steffen Freisinger, Philipp Seeberger, Tobias Bocklet, Korbinian Riedhammer

AI总结:

本文提出利用跨句音频特征进行鲁棒性话题分割,通过多模态方法提升分割性能,并在多个数据集上验证了其有效性。

摘要:

语音内容,如在线视频和播客,往往涵盖多个主题,这使得自动话题分割对用户导航和下游应用至关重要。然而,当前方法未能充分利用声学特征,存在改进空间。我们提出了一种多模态方法,通过微调文本编码器和Siamese音频编码器,捕捉句子边界周围的声学线索。在大规模YouTube视频数据集上的实验显示,与纯文本和多模态基线相比,取得了显著提升。我们的模型还证明了其对ASR噪声的鲁棒性,并在葡萄牙语、德语和英语的三个额外数据集上优于更大的纯文本基线,凸显了学习声学特征在鲁棒性话题分割中的价值。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑