arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1709.02175cs.SD
收藏

归一化特征对提高DNN语音增强泛化能力的影响

Normalized Features for Improving the Generalization of DNN Based Speech Enhancement

Robert Rehr, Timo Gerkmann

摘要:

增强含噪语音是恢复语音质量、提高语音可懂度的一项重要任务。在传统的基于非机器学习(ML)的方法中,降噪所需的参数是从噪声观测中盲估计的,而实际的滤波器函数是基于统计假设解析地导出的。即使这样的方法很好地推广到许多不同的声学条件,瞬态噪声中的噪声抑制能力是低的。为了修正该缺点,已经采用诸如深度学习的机器学习(ML)方法来进行语音增强。然而,由于其数据驱动的性质,基于ML的方法对未知噪声类型的推广仍在讨论中。为了提高基于ML的算法的泛化能力和增强非基于ML的方法的噪声抑制能力,我们提出了两种方法的组合。为此,我们采用先验信噪比(SNR)和后验SNR估计作为输入功能的深度神经网络(DNN)的增强方案。我们表明,这种方法允许基于ML的语音估计器快速推广到未知的噪声类型,即使在训练过程中只看到很少的噪声条件。此外,所提出的功能优于竞争的方法,其中的噪声功率谱密度的估计被附加到嘈杂的频谱。语音质量的感知评估(PESQ)和短时客观可懂度(STOI)等工具性措施表明,当使用所提出的功能时,在看不见的条件下会有很大的改善。听力实验证实了我们提出的组合改进的泛化。

英文摘要:

Enhancing noisy speech is an important task to restore its quality and to improve its intelligibility. In traditional non-machine-learning (ML) based approaches the parameters required for noise reduction are estimated blindly from the noisy observation while the actual filter functions are derived analytically based on statistical assumptions. Even though such approaches generalize well to many different acoustic conditions, the noise suppression capability in transient noises is low. To amend this shortcoming, machine-learning (ML) methods such as deep learning have been employed for speech enhancement. However, due to their data-driven nature, the generalization of ML based approaches to unknown noise types is still discussed. To improve the generalization of ML based algorithms and to enhance the noise suppression of non-ML based methods, we propose a combination of both approaches. For this, we employ the a priori signal-to-noise ratio (SNR) and the a posteriori SNR estimated as input features in a deep neural network (DNN) based enhancement scheme. We show that this approach allows ML based speech estimators to generalize quickly to unknown noise types even if only few noise conditions have been seen during training. Further, the proposed features outperform a competing approach where an estimate of the noise power spectral density is appended to the noisy spectra. Instrumental measures such as Perceptual Evaluation of Speech Quality (PESQ) and short-time objective intelligibility (STOI) indicate strong improvements in unseen conditions when the proposed features are used. Listening experiments confirm the improved generalization of our proposed combination.

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。

↑