Speech Enhancement with a Generative Adversarial Network

By Mira Lilleholt Vik. 2019

104 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Szu-Wei Fu
2. Yu Tsao
3. Kaizhi Qian
4. Yang Zhang
5. Chien-Feng Liao
6. Shiyu Chang
7. Shou-De Lin
8. Loïc Simon

Citations in the corpus

• MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement (2019, methodology): “To achieve this, G is trained to map noisy speech x to its corresponding clean speech y by minimizing the following loss function (as in (Pascual et al., 2017).”
• Revisiting Precision and Recall Definition for Generative Model Evaluation (2019, background): “While generative models have seen successful applications far beyond just image data (such as speech enhancement (Pascual et al., 2017), text to image synthesis (Reed et al., 2016) or text translation (Lample et al., 2018)), we will focus on image generation, as is popular in the recent literature.”
• Zero-Shot Voice Style Transfer with Only Autoencoder Loss (2019, background): “Also, although there is an increasing number of works that introduce GAN to speech generation (Donahue et al., 2018) and speech domain transfer (Pascual et al., 2017; Subakan & Smaragdis, 2018; Fan et al., 2018; Hosseini-Asl et al., 2018), there is no strong evidence that the generated speech…”