Abstract
This work presents a framework based on feature disentanglement to learn speaker embeddings that are robust to environmental variations. Our framework utilises an auto-encoder as a disentangler, dividing the input speaker embedding into components related to the speaker and other residual information. We employ a group of objective functions to ensure that the auto-encoder's code representation - used as the refined embedding - condenses only the speaker characteristics. We show the versatility of our framework through its compatibility with any existing speaker embedding extractor, requiring no structural modifications or adaptations for integration. We validate the effectiveness of our framework by incorporating it into two popularly used embedding extractors and conducting experiments across various benchmarks. The results show a performance improvement of up to 16%. We release our code for this work to be available this https URL
Abstract (translated)
本文提出了一种基于特征分解来学习对环境变化具有鲁棒性的说话人嵌入的工作框架。我们的框架利用自动编码器作为分解器,将输入说话人嵌入分解为与说话人相关的残留信息和其他残留信息。我们采用一组目标函数来确保自动编码器的代码表示 - 被用作精细嵌入 - 只压缩说话人的特征。我们通过其与任何现有的说话人嵌入提取器兼容性来展示我们框架的多样性。我们通过在两个流行的嵌入提取器中集成我们的框架并对其进行各种基准测试来验证我们框架的有效性。结果表明,我们的框架性能提高了16%。我们将我们的代码公开发布在以下网址:https:// URL。
URL
https://arxiv.org/abs/2406.14559