https://github.com/bruceunx/tts-clone-inference

TTS with voice clone
https://github.com/bruceunx/tts-clone-inference

Last synced: 6 days ago
JSON representation

TTS with voice clone

• Host: GitHub
• URL: https://github.com/bruceunx/tts-clone-inference
• Owner: bruceunx
• Created: 2024-09-11T03:41:28.000Z (2 months ago)
• Default Branch: main
• Last Pushed: 2024-09-14T11:32:21.000Z (2 months ago)
• Last Synced: 2024-09-14T21:54:04.582Z (2 months ago)
• Language: Python
• Size: 117 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# TTS-Clone-Inference: Clean Inference and Performance Improvements

This repository contains a modified version of a TTS from coqui-ai that focuses on optimizing the inference pipeline for **clean output** and **improved performance**.

    - parse inference pipeline without training or other parts

## Installation

    - use poetry to install dependencies

## Download models

- download model from `https://huggingface.co/bruceunx/tts-clone-inference/tree/main`

## change code in `main.py`

```python

tts = TTS(model_path="models", config_path="models/config.json", gpu=False)

tts.tts_to_file(text="hello world",

                file_path="./tmp/sample.wav",

                speaker_wav="./tmp/output.wav",

                enable_text_splitting=True,

                language="en")

```

## TODO

- [ ] refactoring code

- [ ] improve performance