arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.01545cs.SDcs.AIcs.CLeess.AS
收藏

利用动态随机扰动实现领域自适应语音增强的有效噪声感知数据仿真

Effective Noise-aware Data Simulation for Domain-adaptive Speech Enhancement Leveraging Dynamic Stochastic Perturbation

National Taiwan Normal University(国立台湾师范大学);Academia Sinica(中央研究院);United-Link Co., Ltd.(联科股份有限公司)

Chien-Chun Wang, Li-Wei Chen, Hung-Shin Lee, Berlin Chen, Hsin-Min Wang

AI总结:

本文提出一种利用噪声编码器和GANs的噪声感知数据仿真方法,通过动态随机扰动增强模型对未知噪声的泛化能力,在VoiceBank-DEMAND上优于现有强基线。

摘要:

跨领域语音增强(SE)通常面临严峻挑战,原因在于未见目标领域中噪声和背景信息的稀缺,导致训练条件与测试条件之间出现不匹配。本研究提出一种新颖的数据仿真方法来解决这一问题,该方法利用噪声提取技术和生成对抗网络(GANs),并且仅需有限的目标带噪语音数据。值得注意的是,我们的方法采用噪声编码器从目标领域数据中提取噪声嵌入。这些嵌入能够恰当地引导生成器合成在声学上适配目标领域的语音,同时真实地保留输入干净语音的语音内容。此外,我们引入了动态随机扰动的概念,它可以在推理过程中向噪声嵌入注入受控扰动,从而使模型能够很好地泛化到未见过的噪声条件。在VoiceBank-DEMAND基准数据集上的实验表明,我们的领域自适应语音增强方法优于现有的基于数据仿真的强基线方法。

英文摘要:

Cross-domain speech enhancement (SE) is often faced with severe challenges due to the scarcity of noise and background information in an unseen target domain, leading to a mismatch between training and test conditions. This study puts forward a novel data simulation method to address this issue, leveraging noise-extractive techniques and generative adversarial networks (GANs) with only limited target noisy speech data. Notably, our method employs a noise encoder to extract noise embeddings from target-domain data. These embeddings aptly guide the generator to synthesize utterances acoustically fitted to the target domain while authentically preserving the phonetic content of the input clean speech. Furthermore, we introduce the notion of dynamic stochastic perturbation, which can inject controlled perturbations into the noise embeddings during inference, thereby enabling the model to generalize well to unseen noise conditions. Experiments on the VoiceBank-DEMAND benchmark dataset demonstrate that our domain-adaptive SE method outperforms an existing strong baseline based on data simulation.

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。

↑