Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dreamerc/twitch-tts
Twitch Text-To-Speech Tool
https://github.com/dreamerc/twitch-tts
python3 pytorch tts twitch
Last synced: 23 days ago
JSON representation
Twitch Text-To-Speech Tool
- Host: GitHub
- URL: https://github.com/dreamerc/twitch-tts
- Owner: dreamerc
- License: mit
- Created: 2022-08-01T07:18:46.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-14T05:34:51.000Z (27 days ago)
- Last Synced: 2025-01-14T06:33:20.553Z (27 days ago)
- Topics: python3, pytorch, tts, twitch
- Language: Python
- Homepage:
- Size: 17.6 KB
- Stars: 7
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# twitch-tts
Twitch Text-To-Speech Tool
This tool is added __!tts__ command, and play audio from Python TTS.
It's Free (twitch-tts under MIT License).
The Goal :
1. Help that some people don't want to use their real voice.
2. Accessible feature shouldn't charge.
3. Making the datasets/models for public domain.
__Use it carefully, it's an early version.__
__Upstream, Coqui.ai is under unknown status.__
Development On Ubuntu Linux.
A. Classical (support CUDA)
```
Twitch.py client - !tts [something] -> Coqui.ai TTS Feature - Curl -> ALSA:aplay
```
B. New (CPU only)
```
Browser (twitch-js) <-> Docker with Coqui.ai TTS
```
## Quick Start (B)
```sh
docker run -it --rm -p 5003:8000 -p 5002:5002 -e TWITCHCHANNEL= dreamerwolf/twitch-tts-server:latest /root/twitch-tts/run-cpu-docker.sh
```
- Open [http://localhost:5003](http://localhost:5003)
- Allow auto play media
- Allow JavaScript
- Wait the Model Download
## Install Guide (A)
---
0. Python Venv and Git
```sh
sudo apt install python3-venv git
mkdir -p ~/python3-venv-root/twitch-tts-venv
python3 -m venv ~/python3-venv-root/twitch-tts-venv
cd ~/python3-venv-root/twitch-tts-venv
source ~/python3-venv-root/twitch-tts-venv
git clone https://github.com/dreamerc/twitch-tts
```
1. Install https://github.com/coqui-ai/TTS and others
```bash
pip install -r requirements.txt
```
2. Start TTS WSGI server , and it will take minutes for the model based on your network.
CPU :
```sh
./run-cpu.sh
```
nVidia CUDA :
```sh
./run-cuda.sh
```
3. Edit config.txt and Start the twitch chatbot.
- Channel & nickname : your twitch account name in URL, but channel name start with #
- OAuth : https://twitchapps.com/tmi/
- client_id & client_secret : https://dev.twitch.tv/console/apps/create
Start :
```bash
python twitch-tts.py
```
## How to Use?
Usage :
use it in chatroom
```
!tts
```
## FAQ
- Q: If you don't want to use browser as a client, you still can use python client.
A: Use [Python venv feature](https://docs.python.org/3/library/venv.html) to install.
- Q: Where is autoplay and JavaScript setting?
A: https://blog.mozilla.org/en/uncategorized/block-autoplay/ (But set allow for Firefox), Chrome is default enabled.