Utilizing a human-perception-related objective function to train a speech
enhancement model has become a popular topic recently. The main reason is that
the conventional mean squared error (MSE) loss cannot represent auditory
perception well. One of the typical hu-man-perception-related metrics, which is
the perceptual evaluation of speech quality (PESQ), has been proven to provide
a high correlation to the quality scores rated by humans. Owing to its complex
and non-differentiable properties, however, the PESQ function may not be used
to optimize speech enhancement models directly. In this study, we propose
optimizing the enhancement model with an approximated PESQ function, which is
differentiable and learned from the training data. The experimental results
show that the learned surrogate function can guide the enhancement model to
further boost the PESQ score (in-crease of 0.18 points compared to the results
trained with MSE loss) and maintain the speech intelligibility.

通过一个可微的近似 PESQ 函数训练语音增强模型，可以提高 PESQ 得分（相对于用 MSE loss 训练的结果增加了 0.18 分），同时保持语音可懂性。

使用已学习的损失函数进行学习：利用 Quality-Net 进行语音增强以提高语音质量的感知评估

Learning with Learned Loss Function: Speech Enhancement with Quality-Net  to Improve Perceptual Evaluation of Speech Quality

Improving speech system performance in noisy environments remains a
challenging task, and speech enhancement (SE) is one of the effective
techniques to solve the problem. Motivated by the promising results of
generative adversarial networks (GANs) in a variety of image processing tasks,
we explore the potential of conditional GANs (cGANs) for SE, and in particular,
we make use of the image processing framework proposed by Isola et al. [1] to
learn a mapping from the spectrogram of noisy speech to an enhanced
counterpart. The SE cGAN consists of two networks, trained in an adversarial
manner: a generator that tries to enhance the input noisy spectrogram, and a
discriminator that tries to distinguish between enhanced spectrograms provided
by the generator and clean ones from the database using the noisy spectrogram
as a condition. We evaluate the performance of the cGAN method in terms of
perceptual evaluation of speech quality (PESQ), short-time objective
intelligibility (STOI), and equal error rate (EER) of speaker verification (an
example application). Experimental results show that the cGAN method overall
outperforms the classical short-time spectral amplitude minimum mean square
error (STSA-MMSE) SE algorithm, and is comparable to a deep neural
network-based SE approach (DNN-SE).

利用条件生成对抗网络（cGANs）从噪声语音的谱图到增强副本，将其应用于语音增强（SE）中，并以感知语音质量（PESQ）、短时客观可懂度（STOI）和说话人验证等方面进行评估。实验结果显示，cGAN 方法整体上优于传统的 STSA-MMSE SE 算法，并且与基于深度神经网络的 SE 方法 (DNN-SE) 相当。