I built a text to speech API with voice cloning n RapidAPI, looking for feedback

Hey, I’ve been working on a small text-to-speech API as a side project.
It supports multiple built-in voices and voice cloning from a reference audio URL.
The API returns raw audio bytes directly, so you can play or save the output without extra steps.

I’m mainly sharing it to get feedback from other developers and see how people would use something like this.

Happy to answer questions or improve things based on suggestions.
You can find it here