[HN Gopher] Show HN: Open-source AI video editor
___________________________________________________________________
Show HN: Open-source AI video editor
Hey HN community! I'm one of the lead devs of this project at
fal.ai and we created an open source lightweight video editor
powered by the latest media AI models. The main goal was to tackle
some challenges when dealing with complex media handling and
encoding on the browser. It all started as an internal experiment
but as we tackled some of the issues it was clear there could be
some value sharing it with the open source community. Some of the
key points and tech stack details: - It uses IndexedDb, so all
data is local (i.e. no auth, no cloud db) - Multiple AI models for
video, image, music and voice-over. APIs are provided by fal.ai -
Built with the typical React+Next.js, Shadcn front-end - Used
remotion.dev for the realtime video preview (this is such a great
project, without it the codebase would be twice as large) - File
uploads so you can bring your own media by uploadthing.com -
ffmpeg for encoding the final video and also some ui tricks, like
the audio waveform We deployed a version of it and for now it's
free to use. We do plan to add some rate limiting and a bring your
own API Key next, but it's open source and I'm curious about what
the community will build on top of it, or derive from it. Customize
your own video app and if you do, please share. If you have any
questions, hit me up!
Author : drochetti
Score : 249 points
Date : 2025-01-23 18:34 UTC (1 days ago)
(HTM) web link (github.com)
(TXT) w3m dump (github.com)
| oldgradstudent wrote:
| Cool project, I played with it a bit a few hours ago.
|
| A small bug report: if you have both a still image and a video in
| the timeline, the video does not play in the preview.
| aykutkardas wrote:
| Glad you liked it. And thanks for reporting, I'll look into the
| issue.
| asadm wrote:
| this looks great! I would like standard timeline scrup, drag-drop
| etc.
|
| Some eta on how long a generation will likely take.
| drochetti wrote:
| Thanks for the feedback, eta would be great indeed. I'll look
| into it.
| kfarr wrote:
| Second on timeline scrub (move timeline position depending on x
| position of click on timeline) and drag and drop asset from
| media into timeline. Also spacebar is usually a basic universal
| shortcut to play/pause
| drochetti wrote:
| Drag and drop the media to the timeline, and drag the media
| along the timeline track is already supported.
|
| We will keep improving the UI, including shortcuts. Thanks a
| lot for the feedback.
| voiper1 wrote:
| Wow, that's so fun. Just upload and generate assets super
| easily... queue the next one while it's working.
| peppertree wrote:
| Open source part is just a thin layer of UI. Built to peddle
| their closed backend, which is built on others' open source work.
| Nice.
| echelon wrote:
| Fal wants to sell compute, and since they're straggling behind
| the compute-as-an-API leaders they've attached their cart to
| the art, image, and video subset of the market.
|
| They think they've got a leg up on their model partners. That
| the foundation model companies will mostly become generic
| copies of one another, servants of the compute layer. There are
| so many foundation video models now, and they'll battle it out
| over dwindling margins. Pika, Runway, Kling - they're all the
| same. And there's also growing open source foundation models.
|
| The thing that stands in Fal's way is that the future of AI
| video for artists is local. Hunyuan and Comfy can run on
| desktop machines, and real artists hate the SaaS model and non
| local stuff. It doesn't look like we'll even need A100s to
| reach Pixar levels of quality.
|
| The ones to watch in the art space are Comfy and Invoke. And
| Adobe.
|
| Fal probably has a future in powering the API-driven YouTube
| Shorts slop, though there's probably an upper limit to the
| number of talking head startups.
|
| But there's no way they win artists over to cloud.
| Sophisticated tools are going to be local. Any SaaS art tools
| that do achieve escape velocity will buy and manage their own
| compute and won't go through Fal as a reseller.
| varispeed wrote:
| Call me thick, but it's like back in the day, rich bored
| person goes to an artist and says "paint me this and that".
| Except now the "rich" person calls themselves an artist and
| "actual" artist is replaced by AI contraption that basically
| regurgitates real artist works. It's kind of comical.
|
| But also makes sense. If you wanted e.g. music, you could
| either look at catalogue of what's out there or try to make
| your own. But if no one is making what tickles you and you
| yourself have no talent to make what you would like to
| listen, the AI comes to rescue.
| dutchbookmaker wrote:
| This is all nonsense.
|
| People don't want to watch two chess computers play even if
| they can both smoke Magnus Carlsen.
|
| Art is an even more extreme version of this. Part of what
| makes a Dali painting a Dali painting is the mimetic desire
| of the general weirdness of the human Salvador Dali.
|
| AI can't displace human mimetic desire. AI art is just
| worthless muzak playing in the background at Dennys. Non-
| artistic/non-creative people right now are amazed at how
| they can make muzak but muzak is shit by its very nature.
| varispeed wrote:
| I think you are correct too. In my opinion both takes are
| correct.
| drochetti wrote:
| That's fair. But I know a lot of smart folks out there that
| have trouble building that "thin layer of UI". So if that helps
| them, mission accomplished.
|
| Anyone can replace the AI layer with their own local models,
| other services... whatever suits your use case and preferences
| is fair game.
| lastdong wrote:
| Thank you for open sourcing the project, the UI looks fantastic.
| I noticed that several comments have mentioned the potential to
| improve functionality by allowing users to replace the backend
| with local models, such as Hunyuan. Would you be open to
| considering this feature in the future? Additionally, if other
| developers were to contribute, do you think this could be a
| possibility?
| drochetti wrote:
| Absolutely. The idea of being open source with a permissive
| license is that we're encouraging anyone to do whatever fits
| their use case.
|
| You can replace anything, deploy on your own server, port it to
| other stacks... whatever brings value to you.
|
| We're also open to PRs, cut an issue in the repo and we can get
| the conversation going.
| Jayakumark wrote:
| Great release , where is the source code for FFMPEG library used?
| SrslyJosh wrote:
| [flagged]
| bdangubic wrote:
| if this comment was a video I'd edit it with this tool :)
| dang wrote:
| Could you please stop posting unsubstantive comments and
| flamebait? Your account has unfortunately been doing this
| repeatedly, e.g.:
|
| https://news.ycombinator.com/item?id=42789451
|
| https://news.ycombinator.com/item?id=42742537
|
| https://news.ycombinator.com/item?id=42742525
|
| That's not what this site is for, and destroys what it is for,
| so we end up having to ban such accounts.
|
| If you wouldn't mind reviewing
| https://news.ycombinator.com/newsguidelines.html and taking the
| intended spirit of the site more to heart, we'd be grateful.
| satvikpendem wrote:
| I was just looking for something like this, is this similar to
| RunwayML and their editor?
| popalchemist wrote:
| Wow, very impressive. Is there any way to use the video editor UI
| component in a package, separately from the rest of this project?
| drochetti wrote:
| Thanks! Not as a component, but you can clone or copy it and
| modify as needed.
|
| It's a quite complex UI, so it's not easy to export it as a
| single component.
| newusertoday wrote:
| you can use remotion which this project is using
| dylan604 wrote:
| Am I missing a demo video or anything, or is the only way to see
| it to install and run it? The only thing I'm seeing is the poster
| image with what I'm assuming is a very cropped image of part of
| the UI???
|
| Just seeing the interface gives a lot of clues as to potential
| usability as if it's targeted at novices or hardcore editors. Not
| being able to see if there's any kind of lag when trying to
| scrub/navigate the video is also something that tells me if I'm
| going to get upset with it from being web based or if it feels
| native.
|
| Just food for thought on what people might be looking for instead
| of having to install something that might not be right for them.
| thangngoc89 wrote:
| It's here: https://fal-video-studio.vercel.app/
| dylan604 wrote:
| and you found that link how? am I just missing it in TFA?
|
| Edit: What do the buttons next to the play/pause button do?
| Typically, I'd expect to frame advance, or jump to next the
| next clip. I click and see nothing happening. I dropped two
| images I uploaded into the timeline, yet there's nothing
| visible in the timeline. The monitor shows something, but
| nothing in the timeline indicating how the clip fits in the
| timeline. There's not indication of what the frame size of
| the timeline is. The images look like they are being
| stretched/squeezed to fit, but there's also no indication in
| the media bin about the media like duration/size/etc.
|
| This really feels targeted for people with no previous
| knowledge of a video editor. It also just has that feel of
| "do something quick to make a post with it's AI to HN".
| drochetti wrote:
| The link is on the GitHub repo.
|
| As for the feedback, fair enough. As I said, this is just a
| demo and it's in its early stages. It has no goal of
| replacing professional video editors or even matching their
| features. Some annoying bugs and missing features, we will
| tackle them.
|
| As for your last comment, feel free to judge, but I can say
| it was not quick at all to do this just to post on HN. You
| can check all the commit history and the codebase so you
| can have an idea how much care was put into it.
| dylan604 wrote:
| It is quite clear by the result that you've spent all of
| your time/effort building some front end to genAI vs
| making a video editor. As of now, I see very little that
| looks like a video editor. There is no basic transport
| controls, there's no obvious way of trimming clips,
| there's nothing that says "you can edit videos" in this.
|
| This is a very premature demo that does more harm than
| good in promoting your product. If you said we built a
| tool that makes it easy to generate content, then you'd
| have a much more interesting product. Tacking on claims
| of this being a video editor is extremely disappointing,
| as the most basic of abilities of a video editor are
| missing. This is what brings on the criticism of being a
| rushed product launch just for the PR postings.
| nikolayasdf123 wrote:
| someone please add "AI Video Editing". I don't care about
| generating assets with AI.
|
| scrolling frame by frame through hours of hundreds of tiny
| videos, fixing their exposure, fixing their dropped frames,
| fixing transitions, fixing and matching visual narration to
| textual and audio narration, aligning music to video, making
| "vlog" like templates. there is just so much to be fixed in this
| industry, but everyone is concerned with generating assets.
| that's just not the biggest issue video editors are facing. like
| not at all.
| EyMaddis wrote:
| Because for generation you just have to make a prompt to OpenAI
| and get a VC to throw money at you. Why bother with the hard
| thing
| nikolayasdf123 wrote:
| you don't even need VC. people would love it and more than
| happy to pay for it, high price! I bet China has already some
| of such apps. I see many very similar vlogs in Little Red
| Note, likely edited automatically by some kind of software.
| build it and. ship it
| t0bia_s wrote:
| Absolutely agree. I don't care about generating footages with
| non existing content, just let me help with editing of already
| taken real life footages to speed up post-production!
| burningion wrote:
| I'm working on this! Here's an example using Claude and Model
| Context Protocol: https://www.youtube.com/watch?v=bEDA9lkkGVM
|
| You can sign up at https://www.video-jungle.com/ and I'll send
| an invite over.
| dgreensp wrote:
| Thanks for posting it!
|
| Remotion does look really cool for in-browser video editing.
| People should note that you can't make free-to-use software with
| Remotion. It is a paid product, and you can download and run the
| code for free in very limited circumstances that let you
| experiment with it. You can't use the output of your own software
| built on Remotion for any for-profit purpose, nor can anyone
| else, without a license.
| wanderingmind wrote:
| Please add a dockerfile to generate a docker image that makes it
| easy to setup and test
| atentaten wrote:
| Nice work! It would be cool to be able to do a basic cross fade
| along with fade in fade out with the clips.
___________________________________________________________________
(page generated 2025-01-24 23:02 UTC)