Is it possible to use F5 to generate TTS without providing an audio/transcript sample?

[Fl4m3Ph03n1x]

I have a few scripts for lectures and I need to find an AI to TTS them into audio files.
I understand F5 is great at cloning voices, but I don't want to clone any voice. I do need an English male deep voice however. This leads me to the following questions:

1. Is it possible to use F5 without cloning a voice? (i.e., does it have a pre-defined selection of voices I can use?)
2. If not, I understand I can train F5 to read text if I train it using a dataset, correct?
3. If point 1 is "no" and point 2 is "yes", where can I find free voice datasets to use train F5? (I assume they need to be in some special format).

Any help with this questions would be highly appreciated.