Automating Conference Video Post-Production with VEditor
Devcon 8 · 5 Nov 2026 · 4:20
Om Vanwari, an open-source contributor at FOSSASIA, presents VEditor, a tool that speeds up post-production for conference talks, which the talk says can take two weeks. An organizer creates a talk, uploads a recording, approves that it matches (gate one), and chooses a generated or uploaded intro and outro. The speaker receives a magic link to a studio to set start and end points and preview the cut, gate two approves it, and the pipeline assembles intro, talk and outro and transcodes the video on a single server instead of personal laptops. Hosting it for upcoming events is the next milestone.
Speakers
Topics
From Devcon 8, hosted by Ethereum Foundation.
Transcript
[music] [music] >> Yeah. Uh so, hello everyone. My name is Om Manwari and I am excited to uh showcase you Veditor. So, if you have ever organized a event like Open Tech Summit India, you know the post-production nightmares. Uh hours of recordings, multiple rooms, multiple days, multiple rooms, multiple days, and oh it it becomes very hard like it takes 2 weeks for uh to process a whole uh talk and get it published ready.
So, here comes here comes Veditor for the solution. What it does is it uh provides automated collaborative way to uh create a published talk with a intro, the talk, and the outro. How does this work? Yeah. So, how does this work is organizer creates a talk, then uploads uh the recording. As you can see, the first step is detect and the video gets detected and the video gets detected and the organizer uh checks whether the video they uploaded is correct or not. If it is correct and it matches the talk, uh so they approve it. That's the gate one approval. Next comes There
comes a There comes a uh step where they they can choose to whether uh get a intro or outro. Whether it can be generated or uploaded by themselves. Yeah. There comes the speaker handoff. A mail gets sent to the speaker with a magic link where they click on it and a studio opens where the speaker has to give a start and a end. Uh and then get a preview generated as you can see. And after the preview generated, the uh they take a look if they're happy, they're satisfied with it. There comes gate two, the review. They can approve it and move forward
with the next transcode part. The pipeline takes in the cut uh cut end points, the start and end. And depending on whether uh the organizer chose to uh start give a uh intro or outro generated, the pipeline assembles the intro, the cut, the outro, and transcodes it, and gives a final finish to the final finish of a published talk. Let's see how it works. Oh.
I think it's not working. Uh but the the uh like how is this different from Oh. Uh Okay. Okay. Oh. The How is this uh making it better? Okay. So, the old way was very tedious. You It was a time-consuming process, and personal laptops were used. It took it take a personal personal laptop's RAM, which was taking a lot of RAM to transcode the whole video, and ev- every laptop should would have the intro, the outro, and everything.
Here, it is a single server where it is hosted, and everything is automated based on uh the cut bounds provided by the speaker. What are the next milestone? To get it hosted and let it be used to on uh the next events. So yes I'm done with my talk and uh yep, if you have any questions and if you would like feel free to reach out. Here's the QR code and uh yep, I'll be outside.
>> [music]