Video wird geladen...
Video konnte nicht geladen werden
This entire video was generated with a single prompt using Vidrush. I just picked a face and asked it to produce a video about SpaceX, from its failures to its IPO. Vidrush handled everything. Research, script, voiceover, visuals, edits and the AI avatar. 19 minutes of finished video, done... show more
10,566 Aufrufe • vor 1 Monat •via X (Twitter)
24 Kommentare

YouTube solved

And you're not stuck with the first output. Every video opens in a real timeline editor. Cut, swap, re-order, or just tell the agent what to fix in plain English.

Start producing videos today → Skip the waitlist with code VIDRUSH

Ve sonra kanal kapanır heygen kanalları gibi

AI avatar kurallara aykırı değil. O kapanan kanallar doktor, avukat vs taklit edenler için. Vidrush zaten izin vermiyor o tarz videolara.

Wow! Exactly just what I was looking for 🤝🏾🔥

cool

Absolutely. A amazing tool

Vidrush is just 🔥

Is it out now?

yes skip the waitlist with the code VIDRUSH

How do you pick face there is no option to pick a face in vidrush

There is. We just released 2 days ago!

looks like it was captured on 60fps, the blur is much, may be you can correct that and it's actually a better output

Vidrush I kodlayan kişinin Türk olduğunu bilmiyordum hep başkasını görüyorduk

3 kurucu ortaktan 2si Türk :)

Looks really great ! Did a similar thing for a Yoder channel Imo next low hanging fruit is doing R&D around how to make these appear more natural. I'm sure team has worked at some levels on it. Had some success at making them a bit more natural but just the talking faces seem 🤖

Nineteen minutes from one prompt is a production feat and a fact-checking liability. I would expose the research manifest beside the video: every factual claim, its source, and whether a human approved it. The first business that can make long-form automation cheap without making corrections expensive will have the stronger product.

you can also provide your own custom script which you would have full control over the narrative.

A custom script solves narrative control, but factual control still needs a separate layer. The useful workflow would lock the approved script, attach a source to every factual sentence, then flag any voiceover or visual that drifts from it during generation.

the research step is the one id check twice. documentary style is all specifics, and thats exactly where it will confidently invent a date

you can provide your own script where you control the full narrative!

true, but the demo is selling the other path. anyone who buys off a one prompt video is going to use the one prompt

really cool, can it also give me an Xlm file so that I can open this into AE and add some on talking head screen motion graphics?
