45 lines
1.5 KiB
Markdown
45 lines
1.5 KiB
Markdown
# Umamusume TTS
|
|
Convert your text into speech using your favorite UmaMusume character's voice.
|
|
|
|
## Prerequisites
|
|
### Install the required packages :
|
|
|
|
Run the following command to install dependencies:
|
|
```
|
|
pip install -r requirements.txt
|
|
```
|
|
*Note: Some packages may not be compatible with certain systems. If you encounter issues, remove the incompatible packages from `requirements.txt` and try again.*
|
|
|
|
## Running the API
|
|
|
|
To start the API server, run:
|
|
```
|
|
python app.py
|
|
```
|
|
|
|
## Testing the API :
|
|
|
|
### Text-to-Speech (TTS) API
|
|
Send a POST request to the `/synthesize` endpoint with the following payload:
|
|
```
|
|
curl -X POST -H "Content-Type: application/json" \
|
|
-d '{
|
|
"speaker_name": "Rice Shower",
|
|
"text": "ライスね、、お兄様のこと、だーーい好き！",
|
|
"noise_scale": 0.37,
|
|
"noise_scale_w": 0.46,
|
|
"length_scale": 1.3
|
|
}' \
|
|
http://localhost:18343/synthesize --output output.ogg
|
|
```
|
|
The response will be an audio file saved as `output.ogg`.
|
|
|
|
### Get Available Speakers
|
|
Fetch a list of available speakers with an optional search term:
|
|
```
|
|
curl -v -X POST http://localhost:18343/speakers -H "Content-Type: application/json" -d '{"search": "mejiro"}'
|
|
```
|
|
|
|
## Notes
|
|
- Ensure your Python environment is properly set up with `pip` and other dependencies before running the application.
|
|
- Modify the `config_path` and `checkpoint_path` in `app.py` to point to the correct configuration and model files if they are not in the default locations. |