arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.06136cs.SDeess.AS
收藏

DENSE:动态嵌入因果目标语音提取

DENSE: Dynamic Embedding Causal Target Speech Extraction

Yiwen Wang, Zeyu Yuan, Xihong Wu

AI总结:

针对静态嵌入难以捕捉已提取语音上下文的问题,提出DENSE动态嵌入因果TSE模型,通过自回归生成上下文相关嵌入,实现实时帧级提取并提升STOI与SDR。

摘要:

目标语音提取(TSE)专注于从混合信号中提取特定目标说话人的语音。现有TSE模型通常利用静态嵌入作为提取目标说话人声音的条件。然而,静态嵌入往往无法捕捉所提取语音信号的上下文信息,这可能会限制模型的性能。为解决这一局限,我们提出了一种新颖的动态嵌入因果目标语音提取模型。我们的方法引入自回归机制,基于已提取的语音生成依赖于上下文的嵌入,从而实现实时、帧级别的提取。实验结果表明,所提出的模型提升了短时客观可懂度(STOI)和信号失真比(SDR),为挑战性场景中的目标语音提取提供了一种有前景的解决方案。

英文摘要:

Target speech extraction (TSE) focuses on extracting the speech of a specific target speaker from a mixture of signals. Existing TSE models typically utilize static embeddings as conditions for extracting the target speaker's voice. However, the static embeddings often fail to capture the contextual information of the extracted speech signal, which may limit the model's performance. We propose a novel dynamic embedding causal target speech extraction model to address this limitation. Our approach incorporates an autoregressive mechanism to generate context-dependent embeddings based on the extracted speech, enabling real-time, frame-level extraction. Experimental results demonstrate that the proposed model enhances short-time objective intelligibility (STOI) and signal-to-distortion ratio (SDR), offering a promising solution for target speech extraction in challenging scenarios.

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。

↑