Credit goes to the developers of Silero TTS
Silero PyTorch Page
Silero GitHub Page
This is primarily to serve the TTS extension in SillyTavern. The TTS module or server can be used any way you wish.
pip install silero-api-server
or, when using venv
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
to leave the virtual environment, use the deactivate command.
python -m silero_api_server will run on default ip and port (0.0.0.0:8001)
usage: silero_api_server [-h] [-o HOST] [-p PORT]
Run Silero within a FastAPI application
options:
-h, --help show this help message and exit
-o HOST, --host HOST
-p PORT, --port PORT
-r, --reload Enable automatic reloading on file change
On first run of server, two operations occur automatically. These may take a minute or two.
- The model will be downloaded
- Voice samples will be generated.
API Docs can be accessed from http://localhost:8001/docs
Samples are served statically by the web server at /samples/{speaker}.wav or callable from the API from /tts/sample?speaker={speaker} endpoint.
Download and set the desired language using POST /tts/language with payload {"id":"languageId"}
List of language ids are available via GET /tts/language