Install Manager
Added the option to select which model each AI package uses, with the recommended model selected by default.
Improved performance and compatibility for lower-end PCs.
Improved guidance for systems with limited VRAM, making it easier to choose the most suitable model.
Workflow and Interface
Added Save Last TTS support through both a hotkey and the Quick Bar.
Made additional improvements to the main TTS interface and overall workflow.
Fixes & Stability
Several fixes have also been made throughout the app. I'll continue releasing hotfixes and smaller patches as needed to improve stability.
As always, I'm open to suggestions. If there's something you think would improve your workflow or the way you use the software, let me know.
What's Next
Text to Sing is mostly complete! I've already started working on what comes after it, including speech-to-text related features and a completely new cat overlay model, a much more polished and thoughtfully designed desktop pet.
The software has grown far beyond anything I originally imagined. What started as a tiny tool I made for myself has somehow grown into all of this.
Thank you to everyone using it, sharing feedback, and helping it grow. It's still a little surreal knowing there are actual people out there using this once-tiny, now slightly overgrown tool.