Uses StableDiffusion XL for images by default (more expensive but much higher quality)
Uses ChatGPT 3.5 for scripts by default (higher quality scripts)
New script generation algorithm (much cheaper)
Automatic background music (pulled from freepd.com)
Changed video backend from MoviePy to FFmpeg (much more powerful, flexible. Also, much better-looking subtitles)
Significant codebase refactoring (with even more refactoring on the way)
There are still many unfinished features that I'm still planning to add (e.g., automatic YT uploads). But I'm releasing this as V2 right now since has feature parity with V1 and shouldn't break anything.
Major changes include:
There are still many unfinished features that I'm still planning to add (e.g., automatic YT uploads). But I'm releasing this as V2 right now since has feature parity with V1 and shouldn't break anything.