Best Open source TTS right now for narration?

I run these models on Kaggle notebook, so not all TTS models, such as the ones that use conda env, are compatible (Or I just haven't found a way for them to work on Kaggle). I currently use a fork from Chatterbox called Chatterbox Audiobook. It is like a workstation really optimized for getting the close-to-perfection audio clips from Chatterbox. However, the only downside of Chatterbox is the lack of emotional sliders or tags that you can use to control the output. Chatterbox Turbo seems to fix that with tags, but it still lacks the range of emotions that you can see from Google Gemini TTS. However, the problem with Google Gemini is that the voice sounds different for each generation, which can't be fixed even with RVC. Looking at the current leaderboard, Breeze TTS is something I have never tried but am unsure due to its description, which seems to be tailored for mainly realtime stuff. What are the current must-try options for audiobook narration? The features I am looking for include voice cloning, emotional tags, and natural speech. Much appreciated for your input.