arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2506.02499cs.SDeess.AS
收藏

DnR-nonverbal:包含非语言声音的电影音频源分离数据集

DnR-nonverbal: Cinematic Audio Source Separation Dataset Containing Non-Verbal Sounds

LY Corporation(LY公司)

Takuya Hasumi, Yusuke Fujita

AI总结:

针对电影音频源分离中笑声、尖叫等非语言人声被误分为音效的问题,本文构建 DnR-nonverbal 数据集,并验证其可提升合成与真实电影音频中的分离效果。

摘要:

我们提出了一个用于电影音频源分离(CASS)的新数据集,用于处理非语言声音。现有 CASS 数据集仅包含朗读风格的声音作为语音声源。这些数据集与真实电影音频不同,后者更可能包含表演出来的人声。因此,在传统数据集上训练的模型往往会出现问题:笑声和尖叫等情绪强烈的声音更容易被分离为音效,而不是语音。为了解决这个问题,我们构建了一个新数据集 DnR-nonverbal。所提出的数据集在语音声源中包含笑声和尖叫等非语言声音。通过实验,我们揭示了当前 CASS 模型在非语言声音提取方面的问题,并表明我们的数据集能够在合成音频和真实电影音频中有效解决该问题。我们的数据集可在 https://zenodo.org/records/15470640 获取。

英文摘要:

We propose a new dataset for cinematic audio source separation (CASS) that handles non-verbal sounds. Existing CASS datasets only contain reading-style sounds as a speech stem. These datasets differ from actual movie audio, which is more likely to include acted-out voices. Consequently, models trained on conventional datasets tend to have issues where emotionally heightened voices, such as laughter and screams, are more easily separated as an effect, not speech. To address this problem, we build a new dataset, DnR-nonverbal. The proposed dataset includes non-verbal sounds like laughter and screams in the speech stem. From the experiments, we reveal the issue of non-verbal sound extraction by the current CASS model and show that our dataset can effectively address the issue in the synthetic and actual movie audio. Our dataset is available at https://zenodo.org/records/15470640.

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。

↑