arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2212.05008eess.AScs.SD
收藏

双曲音频源分离

Hyperbolic Audio Source Separation

MERL(三菱电机研究实验室)

Darius Petermann, Gordon Wichern, Aswin Subramanian, Jonathan Le Roux

摘要:

我们介绍了一个框架,音频源分离使用嵌入在双曲流形上,competitively表示声源和时频特征之间的层次关系。受最近成功地用双曲嵌入对文本和图像中的层次关系进行建模的启发,我们的算法为混合信号的每个时频点获得双曲嵌入,并使用双曲softmax层估计掩码。在一个包含多个人说话和乐器演奏的混合物的合成数据集上,我们的双曲模型在源失真比方面表现出与欧几里得基线相当的性能,在低嵌入维数下表现更好。此外,我们发现包含多个重叠源的时频区域被嵌入到中心(即,最不确定的区域),并且我们可以使用该确定性估计来在隔离各个声音时在伪像引入和干扰减少之间有效地进行权衡。

英文摘要:

We introduce a framework for audio source separation using embeddings on a hyperbolic manifold that compactly represent the hierarchical relationship between sound sources and time-frequency features. Inspired by recent successes modeling hierarchical relationships in text and images with hyperbolic embeddings, our algorithm obtains a hyperbolic embedding for each time-frequency bin of a mixture signal and estimates masks using hyperbolic softmax layers. On a synthetic dataset containing mixtures of multiple people talking and musical instruments playing, our hyperbolic model performed comparably to a Euclidean baseline in terms of source to distortion ratio, with stronger performance at low embedding dimensions. Furthermore, we find that time-frequency regions containing multiple overlapping sources are embedded towards the center (i.e., the most uncertain region) of the hyperbolic space, and we can use this certainty estimate to efficiently trade-off between artifact introduction and interference reduction when isolating individual sounds.

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。

↑