High-Logic/Genie-TTS
A lightweight ONNX-based inference engine for GPT-SoVITS text-to-speech models.
Not currently ranked — collecting fresh signals.
star history
This repository provides a text-to-speech inference engine built on GPT-SoVITS, converting PyTorch models to ONNX format for optimized performance. It supports multiple languages (Japanese, English, Chinese, Korean) and achieves faster first inference latency compared to official implementations. The project includes API server functionality and targets CPU-based deployment with reduced runtime size.
Frequently asked
- What is High-Logic/Genie-TTS?
- A lightweight ONNX-based inference engine for GPT-SoVITS text-to-speech models.
- Is Genie-TTS open source?
- Yes — High-Logic/Genie-TTS is open source, released under the MIT license.
- What language is Genie-TTS written in?
- High-Logic/Genie-TTS is primarily written in Python.
- How popular is Genie-TTS?
- High-Logic/Genie-TTS has 1.6k stars on GitHub.
- Where can I find Genie-TTS?
- High-Logic/Genie-TTS is on GitHub at https://github.com/High-Logic/Genie-TTS.