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 2108.07974eess.AS
收藏

FDN:用于与文本无关的说话人确认的分层卷积有限差分网络

FDN: Finite Difference Network with Hierarchical Convolutional Features for Text-independent Speaker Verification

Jin Li, Nan Yan, Lan Wang

摘要:

近年来,使用原始波形作为深度网络的输入已被广泛探索用于说话人验证系统。例如,RawNet和RawNet 2从波形中自动提取说话人的特征嵌入来识别他们的语音,这可以大大减少前端计算并获得最先进的性能。然而,这些模型没有考虑说话人的高级行为特征,如语调,表明每个说话人的通用风格,节奏,\textit{etc}。本文提出了一种新的网络,可以处理语调信息,通过计算不同的发言者的话语变化的有限差分。此外,还设计了一种由粗到细的层次化方法来增强语调特性,以提高系统的准确性。然后,将高级语调特征与低级嵌入特征融合。在官方VoxCeleb 1测试数据、VoxCeleb 1-E和VoxCeleb-H协议上的实验结果表明,该方法优于现有的最先进的系统。为了便于进一步研究,代码可在https://github.com/happyjin/FDN上获得

英文摘要:

In recent years, using raw waveforms as input for deep networks has been widely explored for the speaker verification system. For example, RawNet and RawNet2 extracted speaker's feature embeddings from waveforms automatically for recognizing their voice, which can vastly reduce the front-end computation and obtain state-of-the-art performance. However, these models do not consider the speaker's high-level behavioral features, such as intonation, indicating each speaker's universal style, rhythm, \textit{etc}. This paper presents a novel network that can handle the intonation information by computing the finite difference of different speakers' utterance variations. Furthermore, a hierarchical way is also designed to enhance the intonation property from coarse to fine to improve the system accuracy. The high-level intonation features are then fused with the low-level embedding features. Experimental results on official VoxCeleb1 test data, VoxCeleb1-E, and VoxCeleb-H protocols show our method outperforms and robustness existing state-of-the-art systems. To facilitate further research, code is available at https://github.com/happyjin/FDN

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。

↑