论文标题
复杂STFT域中的基于得分的生成模型的语音增强
Speech Enhancement with Score-Based Generative Models in the Complex STFT Domain
论文作者
论文摘要
基于分数的生成模型（SGM）最近显示了难以生成的任务的令人印象深刻的结果，例如无条件和条件生成自然图像和音频信号。在这项工作中，我们将这些模型扩展到复杂的短时傅立叶变换（STFT）领域，并提出了使用复杂值的深层神经网络来增强语音的新型训练任务。我们在随机微分方程（SDE）的形式主义中得出了这项训练任务，从而实现了预测器 - 矫正器采样器的使用。我们提供了受到先前出版物启发的替代配方，以使用生成扩散模型来增强语音，从而避免了对噪声分布的任何先前假设的需求，并使训练任务纯粹是生成生成的，这是我们所显示的，从而改善了增强性能。
Score-based generative models (SGMs) have recently shown impressive results for difficult generative tasks such as the unconditional and conditional generation of natural images and audio signals. In this work, we extend these models to the complex short-time Fourier transform (STFT) domain, proposing a novel training task for speech enhancement using a complex-valued deep neural network. We derive this training task within the formalism of stochastic differential equations (SDEs), thereby enabling the use of predictor-corrector samplers. We provide alternative formulations inspired by previous publications on using generative diffusion models for speech enhancement, avoiding the need for any prior assumptions on the noise distribution and making the training task purely generative which, as we show, results in improved enhancement performance.
