Hey everyone — new guy here. Thanks for having me! I’ve been following what Scotty has been doing with Keira and some of the other AI characters, and the realism is what really caught my attention. The first time I saw @Keira K , I genuinely thought she was a real person. I’m working on a project where we’re considering creating two recurring virtual hosts — one male and one female — and the quality requirement is very high. They need to look and move convincingly enough that they don’t immediately read as AI, while still being clearly disclosed as virtual/AI hosts. I have a few questions about whether ShortPulse could handle the workflow we have in mind: 1. Is Keira’s current talking-video workflow created primarily or entirely inside ShortPulse? If so, which model/workflow is being used for her facial animation, lip-sync, movement, and overall realism? 2. Can ShortPulse use a downloaded NotebookLM Audio Overview as the audio source? Specifically, NotebookLM can generate a conversation between two speakers. Could we take that two-person audio and use it with two persistent ShortPulse characters? 3. If we have one male character and one female character, can ShortPulse automatically distinguish the two speakers in the audio and assign Speaker A to one character and Speaker B to the other, with each character only speaking their own lines? 4. If it cannot do that automatically, what would be the best ShortPulse workflow? Would we need to separate the two speakers into individual audio tracks first and generate each character separately? 5. Can the same two characters be kept truly consistent across many videos — same face, body, general appearance, voice, and identity — while changing clothing, locations, camera angles, etc.? 6. Can each character have a permanent reusable voice, so they sound like the same two people in every episode? 7. Can ShortPulse create a convincing two-person conversation visually, including alternating close-ups, two-shots, reactions while the other person is speaking, and natural facial/head/body movement? Or would the better approach be to generate each person separately and edit the conversation together afterward? 8. Finally, for people who have done this kind of recurring photorealistic-character work in ShortPulse, which models/workflows are currently giving the most realistic human results? Cost matters, but realism matters much more for this project.

I posted this on the Creator Secrets community board. I'm looking for existing instruction videos or other resources to narrow my experimentation. I'm on the verge of posting my first YouTube video for my new (and currently only) channel, Mindset Business Academy; a channel addressing financial concerns of financially established American seniors 55+. I have a mascot character, Boaz, a wise old owl. I'm soliciting ChatGPT for instructions on how I will overlay a speaking, moving Boaz who narrates my videos, and frequently makes on-screen appearances. I want to generate Boaz's motion clips with motion-matching-voice against a clean background from the start (most AI video generators let me specify this in the prompt) so I can overlay my character on videos as the narrator and instructor. That sidesteps the need for AI rotoscoping entirely and lets me use the simplest, free or paid tool (ShortPulse?, Adobe Premier, DaVinci Resolve, or CapCut, etc.) for the actual compositing. I will save the heavier AI-isolation tools for later if I end up with existing Boaz footage I can't regenerate cleanly. (The attachment is created with ShortPulse).