v1.2.0 Release Notes:
1. Added voice pitch controls and previews.
2. Fixed All tabs playing for same website bug

Release Notes (v1.1.0)

UI Enhancements: The user interface has been polished and updated for a smoother experience.

Piper Neural TTS Integration: Added support for local neural text-to-speech models powered by Piper.

On-Demand Model Downloads: Users can now browse, download, and utilize neural voice models directly within the extension.


In this update, we’ve completely overhauled how TTSFlow handles background navigation and text extraction to make it virtually immune to modern website security, ensuring your hands-free reading never stops.

What's New & Improved:

1. Universal Auto-Play & Memory: TTSFlow now uses Hard Navigation combined with browser local storage. When moving to a new chapter, it securely reloads the page and autonomously re-opens the reader overlay. This completely bypasses Cloudflare bot-protection blocks on highly secure sites like WTR-LAB.
2. Smart URL Predictor: If a website hides its "Next Chapter" links inside complex JavaScript buttons (Single Page Applications like React/Next.js), TTSFlow will now mathematically predict the next chapter's URL and take you there automatically.
3. Ad-Warning Immunity: The text scraper has been upgraded to intelligently ignore injected ad-blocker popups and dummy text wrappers, ensuring it only reads the actual novel content.

Bug Fixes:

1. Voice Selection Sync: Fixed a frustrating browser caching bug where the selected voice model would fall out of sync with the actual audio being played. Your chosen voice is now strictly locked in.
2. UI Polish: Removed the faint circular background artifact when hovering over media controls, keeping the glassmorphism navbar looking sharp, clean, and modern.