Exploring new video formats - feedback needed
Added 2025-07-11 17:00:45 +0000 UTC
Hey folks, I've been thinking that the videos do not have nearly the amount of animation needed to make them feel "alive" and I've been thinking about how to change that.
Check this short video out and tell me what you think.
I feel like it's too early to start researching video creation; I still have a lot to learn about images and other related topics that will be helpful later, and I don't want to skip ahead.
So my idea was this - Align images with voice and add subtitles. This is an early alpha idea. I'll also post a longer video in a bit, but it would be helpful to hear some thoughts. My main concern is that the transitions are too quick and abrupt, which would kill the hypnosis vibe in long-term audios, but again, that's a pure speculation.
P.S.: Getting kind of better with the image generation, so for those who like to be stepped on.
Comments
Let's see if I understand you correctly. The current video format is basically an image slideshow + hypnosis audio on top. The images and the hypnosis are not "connected". What you suggest is, for script sentences like "I look at you and say 'stop moving' and you're immediately frozen in place." The 'stop moving' part, the direct verbal communication, can be specifically used to generate images that fit it. The other thing is, not to try to 100% fit the audio as images? Sometimes, "imperfect" means just some kind of visuals that are generally pleasant to look at (e.g., women). Another thing is to generally include not only women, for example, if there's a chair in the narration, I can generate a chair and show an image of it? And finally, consider adding "sections" to the videos, focusing on visuals, a section with the hypnosis spiral and subtitles, and another with rapid switching images unrelated to the audio. Essentially, think of the video as something more than "linear hypnosis script reading + visuals," but as an experience that incorporates elements of all those to create something more? I probably got stuff wrong, so please elaborate when you have time.
Sect
2025-07-12 11:01:18 +0000 UTCThe test was actively engaging and is nice to focus on and keeps your attention easier compared to static style and I think has promise. In regards to the "Perfect" fit though I think continuity for portions of the video are good like when being spoken to directly by the narrator or for long durations but I'd consider interweaving shorter portions that are "Imperfect" sometimes for different effects. I kind of agree that the interweaving of the images in the demo is a little jarring but its a space where you have a lot of listener attention with an "Expected" visual stimulus and you have opportunities to put something else in depending on the topic so I think it's something that could be used positively with experimentation. Personal opinion but having variation in the images, spirals, text effects, etc so that it doesn't feel like the style repeats too frequently I think would also make it easier to watch. Has my vote.
CheekyBeard
2025-07-12 03:45:55 +0000 UTCReally good feedback! I'm wondering if I can achieve a minimal or smooth camera change, as it would make a noticeable difference, much like treating each image as a frame of a video, as you mentioned. That would be one test to try. Regarding the current visuals, would you happen to have any feedback? I guess that they're either not engaging enough or too repetitive, which can make them boring after 5-10-15 minutes. The hypnosis audio is, in my opinion, the main source of entertainment. But again, I am speculating.
Sect
2025-07-11 19:29:41 +0000 UTCEdit: usually while watching your videos I occasionally close my eyes for a few seconds and open again. I guess that was closely tied to the fact that the images were static for the most part. In this short clip the changing pictures don't seem to be much coherent but it feels like they should be. And that brings some "Uncanny valley" effect. Like, it's neither HMV where there is a new girl on every next slide, but nor it's 5-10fps animation of the same girl either
progamoto
2025-07-11 18:05:10 +0000 UTC
