Differentiable Consistency Constraints for Improved Deep Speech Enhancement
Differentiable Consistency Constraints for Improved Deep Speech Enhancement
复制标题
用于改进深度语音增强的可微一致性约束
DOI：
10.1109/icassp.2019.8682783
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
R. Saurous
中科院分区：
文献类型：
--
作者：
Scott Wisdom;J. Hershey;K. Wilson;J. Thorpe;Michael Chinen;Brian Patton;R. Saurous
近年来，深度网络通过将语音增强视为数据驱动的模式识别问题，使语音增强得到了显著改善。在许多现代增强系统中，大量数据用于训练深度网络来估计复值短时傅立叶变换（STFT）的掩码，以抑制噪声并保留语音。然而，目前的掩蔽方法往往忽略了两个重要的约束：STFT一致性和混合一致性。在没有STFT一致性的情况下，系统的输出不一定是时域信号的STFT，并且在没有混合一致性的情况下，估计的源的总和不一定等于输入混合。此外，之前唯一应用混合一致性的方法使用实值掩码;对于复值掩码，混合一致性已被忽略。
In recent years, deep networks have led to dramatic improvements in speech enhancement by framing it as a data-driven pattern recognition problem. In many modern enhancement systems, large amounts of data are used to train a deep network to estimate masks for complex-valued short-time Fourier transforms (STFTs) to suppress noise and preserve speech. However, current masking approaches often neglect two important constraints: STFT consistency and mixture consistency. Without STFT consistency, the system’s output is not necessarily the STFT of a time-domain signal, and without mixture consistency, the sum of the estimated sources does not necessarily equal the input mixture. Furthermore, the only previous approaches that apply mixture consistency use real-valued masks; mixture consistency has been ignored for complex-valued masks.