A Voice Cloning Method Based on the Improved HiFi-GAN Model

Figure 3

Encoder-decoder model structure.