🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

HiFi-GAN: High Fidelity Denoising and Dereverberation Based on Speech Deep Features in Adversarial Networks

Include Basis-MelGAN, MelGAN, HifiGAN and Multiband-HifiGAN, maybe NHV in the future.

This is an implementation for train hifigan part of XTTSv2 model using Coqui/TTS.

Speech synthesis (TTS) in low-resource languages by training from scratch with Fastpitch and fine-tuning with HifiGan

zero-shot realtime TTS system, fully offline, free and open source

Ultrafast GAN based Vocoder for Text to Speech

HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis

🎙️ Arabic TTS models (FastPitch, Mixer-TTS) in the ONNX format — Python package for offline speech synthesis 🚀📦

RepVgg + HiFiGAN

SA-toolkit: Speaker speech anonymization toolkit in python

Fast and Small codec for flow-matching models

RADTTS + HiFiGAN vocoder

A lightweight HiFi-GAN project for generating realistic speech conditioned on speaker embeddings.

homework for deep generation. Combine FastSpeech2 with different vocoders ⭐REFERENCE (modify origin repos): https://github.com/ming024/FastSpeech2 https://github.com/NVIDIA/waveglow https://github.com/mindslab-ai/univnet https://github.com/jik876/hifi-gan

Stage 1 singing voice conversion with SoftVC-style acoustic modeling, RMVPE pitch conditioning, and HiFi-GAN vocoding.

🎤 Enhance multilingual communication with T5Gemma-TTS, a versatile Text-to-Speech model supporting easy training and inference.

Accelerate Qwen3-TTS inference with Triton kernel fusion for faster text-to-speech performance on NVIDIA GPUs