myshell-ai/OpenVoice ? reverse-engineered prompt

Build me a simple web app for voice cloning. I want to be able to upload a short audio clip of someone's voice, and then type in some text to have the app generate audio of that text being spoken in the cloned voice. The voice clone needs to be very accurate, capturing the original speaker's tone and style.

I'd also like to be able to control the emotion of the voice, like making it sound happy, sad, or excited. Another key feature is that it should work across different languages. For example, I need to be able to provide a voice sample in English and have it generate speech in Spanish, French, Chinese, or Japanese.

Please use Python and the OpenVoice model for this. You can look up its documentation if you need to. Just make a basic interface so I can easily upload the reference voice and type the text.