How to correct the pronunciation of polyphones and output subtitle files with timestamps

[single1225]

I deployed f5-tts locally. When performing speech synthesis, two problems bothered me.
First, I want to correct the pronunciation of Chinese polyphones. Is this function possible? If so, how to implement it in Python code? I haven't seen any relevant documents.
Second, is there any way to output subtitle files with timestamps while outputting audio?

Thank you very much for your help