Best Open source TTS right now for narration?

A user on Kaggle notebooks is seeking an open-source Text-to-Speech (TTS) model for narration, specifically looking for features like voice cloning, emotional tags, and natural speech. They currently use Chatterbox Audiobook but note its lack of emotional control. While Chatterbox Turbo offers tags, it doesn't match the emotional range of Google Gemini TTS, which itself suffers from inconsistent voice generation across outputs, even with RVC.