A standalone implementation of HiFiGAN vocoder for Arabic text-to-speech, based on the paper "HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis" (
https://arxiv.org/pdf/2010.05646.pdf).
1from hifigan_ar_v2 import HiFiGANArabicGenerator
2import torch
3
4# Load the model
5model = HiFiGANArabicGenerator.from_pretrained("generator.ckpt", "config.json")
6
7# Generate audio from mel spectrogram
8mel = torch.rand(1, 80, 122) # Example mel spectrogram
9audio = model(mel) # Shape: [1, 1, 8448]
