[
  {
    "i": 0,
    "speaker": "Speaker 1",
    "text": "Hey everyone, it's dark from cod code here. I heard a lot of questions about how"
  },
  {
    "i": 1,
    "speaker": "Speaker 1",
    "text": "How did fable edits on launch video? So I want to go over that"
  },
  {
    "i": 2,
    "speaker": "Speaker 1",
    "text": "You know, it also made this deck. So it's a little meta. It's also editing this video"
  },
  {
    "i": 3,
    "speaker": "Speaker 1",
    "text": "Yeah, but it did edit this launch video and so how did how did it work?"
  },
  {
    "i": 4,
    "speaker": "Speaker 1",
    "text": "So I worked with a video production crew and we did 17 takes so there were four scenes in it and across that we did"
  },
  {
    "i": 5,
    "speaker": "Speaker 1",
    "text": "a bunch of different takes"
  },
  {
    "i": 6,
    "speaker": "Speaker 1",
    "text": "And this is what I got right like not color graded"
  },
  {
    "i": 7,
    "speaker": "Speaker 1",
    "text": "25 gigabytes worth of files and ended up with this so color graded we had some UI"
  },
  {
    "i": 8,
    "speaker": "Speaker 1",
    "text": "we stitched it all together and cloud code did everything I didn't touch like a"
  },
  {
    "i": 9,
    "speaker": "Speaker 1",
    "text": "Didn't touch it terminal or so I didn't touch a video editor. So"
  },
  {
    "i": 10,
    "speaker": "Speaker 1",
    "text": "Yeah, what does it look like the starting prompt is this I'm processing a bunch of recordings of videos"
  },
  {
    "i": 11,
    "speaker": "Speaker 1",
    "text": "I want you to run a"
  },
  {
    "i": 12,
    "speaker": "Speaker 1",
    "text": "Transcription service on them and stitch together the best shots into one final clip"
  },
  {
    "i": 13,
    "speaker": "Speaker 1",
    "text": "You know notes their multiple takes per scene"
  },
  {
    "i": 14,
    "speaker": "Speaker 1",
    "text": "The best takes are usually the ones at the end with the fewest arms, but that can change"
  },
  {
    "i": 15,
    "speaker": "Speaker 1",
    "text": "I can also I we shot the final scene"
  },
  {
    "i": 16,
    "speaker": "Speaker 1",
    "text": "You know and I asked it to create a JSON file for the video that shows per scene"
  },
  {
    "i": 17,
    "speaker": "Speaker 1",
    "text": "Each clip we're using and so this is all like all I gave it with this folder of videos"
  },
  {
    "i": 18,
    "speaker": "Speaker 1",
    "text": "I didn't tell it for example, which videos are in which scenes, right?"
  },
  {
    "i": 19,
    "speaker": "Speaker 1",
    "text": "And what it did is it transcribed every one of them. So it used whisper and"
  },
  {
    "i": 20,
    "speaker": "Speaker 1",
    "text": "It created this array of words per video and it you know deduped them and chose the best one"
  },
  {
    "i": 21,
    "speaker": "Speaker 1",
    "text": "So I'll be 17 takes and figure out which ones are in which scenes"
  },
  {
    "i": 22,
    "speaker": "Speaker 1",
    "text": "and then it"
  },
  {
    "i": 23,
    "speaker": "Speaker 1",
    "text": "You know like finds the best one for it, right and it compiles all of this into an edited JSON file of"
  },
  {
    "i": 24,
    "speaker": "Speaker 1",
    "text": "And it compiles all of this into an edited JSON file"
  },
  {
    "i": 25,
    "speaker": "Speaker 1",
    "text": "Basically, it has every see in the candidate takes the reasons"
  },
  {
    "i": 26,
    "speaker": "Speaker 1",
    "text": "It chose the the scenes it did and the start and end times per scene"
  },
  {
    "i": 27,
    "speaker": "Speaker 1",
    "text": "if this doesn't do any clipping in the middle of the video, but it totally could have you know, like this is not something that I"
  },
  {
    "i": 28,
    "speaker": "Speaker 1",
    "text": "I tried to make each cut fairly clean. So"
  },
  {
    "i": 29,
    "speaker": "Speaker 1",
    "text": "And then you get F of M peg to like stitch together the first cut. So"
  },
  {
    "i": 30,
    "speaker": "Speaker 1",
    "text": "This is what it looked like. This is actually a little bit postgrading. So"
  },
  {
    "i": 31,
    "speaker": "Speaker 1",
    "text": "This is what it looked like. This is actually a little bit postgrading. So"
  },
  {
    "i": 32,
    "speaker": "Speaker 1",
    "text": "But yeah, yeah, the F of M peg does this"
  },
  {
    "i": 33,
    "speaker": "Speaker 1",
    "text": "Now then we can do the color grader. So the color grader looks muted. Like this is actually what it looked like here in neutral 709"
  },
  {
    "i": 34,
    "speaker": "Speaker 1",
    "text": "What do we want to end up with?"
  },
  {
    "i": 35,
    "speaker": "Speaker 1",
    "text": "and"
  },
  {
    "i": 36,
    "speaker": "Speaker 1",
    "text": "Yeah, how can we regrade it? So it writes, you know, I know nothing about color grading, right?"
  },
  {
    "i": 37,
    "speaker": "Speaker 1",
    "text": "I just know that like it's a thing. So I just asked it to write"
  },
  {
    "i": 38,
    "speaker": "Speaker 1",
    "text": "a bunch of color gradings and"
  },
  {
    "i": 39,
    "speaker": "Speaker 1",
    "text": "This is the final you know off the camera to like to a neutral color grade. So"
  },
  {
    "i": 40,
    "speaker": "Speaker 1",
    "text": "So did that the next point is I"
  },
  {
    "i": 41,
    "speaker": "Speaker 1",
    "text": "Want to create like design assets, so I I gave it previously the script"
  },
  {
    "i": 42,
    "speaker": "Speaker 1",
    "text": "And I had to make a bunch of like design assets and then I said, okay, like I want to turn"
  },
  {
    "i": 43,
    "speaker": "Speaker 1",
    "text": "use Remotion. Remotion is a library that like takes video and react components and allows you"
  },
  {
    "i": 44,
    "speaker": "Speaker 1",
    "text": "to like kind of export them all together into an end video. And so I said, okay, let's make a bunch"
  },
  {
    "i": 45,
    "speaker": "Speaker 1",
    "text": "of components for Remotion, uh, that showed that implements design. So this is what it looks like."
  },
  {
    "i": 46,
    "speaker": "Speaker 1",
    "text": "It's all reacted, like, um, chooses the beats and timings using the transcription JSON that we had"
  },
  {
    "i": 47,
    "speaker": "Speaker 1",
    "text": "before. Right. So it has all this data inside of itself, you know, fades things in and out,"
  },
  {
    "i": 48,
    "speaker": "Speaker 1",
    "text": "um, has a bunch of like, you know, knobs I can adjust. Um, and, uh, the final edit is this sort"
  },
  {
    "i": 49,
    "speaker": "Speaker 1",
    "text": "of like, you know, when to show which, uh, UI component. Right. And so here's an example of,"
  },
  {
    "i": 50,
    "speaker": "Speaker 1",
    "text": "you know, uh, it switching like a UI component on the right beat. So as soon as I say is cloud"
  },
  {
    "i": 51,
    "speaker": "Speaker 1",
    "text": "doing the right work, it switches the two, two components. And again, it's doing this all"
  },
  {
    "i": 52,
    "speaker": "Speaker 1",
    "text": "by itself, right. Based on the transcription, um, and the, and the Remotion component. So"
  },
  {
    "i": 53,
    "speaker": "Speaker 1",
    "text": "uh, then I say like, okay, let's export the designs to Figma. Um, and, uh, yeah, I want to"
  },
  {
    "i": 54,
    "speaker": "Speaker 1",
    "text": "let the design team be able to edit it. Right. And so, um, I start with this, this was like a first"
  },
  {
    "i": 55,
    "speaker": "Speaker 1",
    "text": "pass essentially in like a cloud design like way, but I know the design team has much better taste"
  },
  {
    "i": 56,
    "speaker": "Speaker 1",
    "text": "than me. Um, and I also give them like the ability to edit the color grader. They end up doing this,"
  },
  {
    "i": 57,
    "speaker": "Speaker 1",
    "text": "but like, you know, like I sent them an HTML file that like lets them, uh, play with, uh, some of"
  },
  {
    "i": 58,
    "speaker": "Speaker 1",
    "text": "the, um, uh, like the, the settings. Um, and then, so they, they go edit the Figma and they give"
  },
  {
    "i": 59,
    "speaker": "Speaker 1",
    "text": "it back to me. And I just say, this is all using the Figma MCP. I just say, the design has been"
  },
  {
    "i": 60,
    "speaker": "Speaker 1",
    "text": "updated in this Figma. Can you update the video to match? Um, and it does this, right. So, um,"
  },
  {
    "i": 61,
    "speaker": "Speaker 1",
    "text": "it like turns it into this like clean, uh, like, you know, transparent look that the design team"
  },
  {
    "i": 62,
    "speaker": "Speaker 1",
    "text": "put together. And this is what it looks like."
  },
  {
    "i": 63,
    "speaker": "Speaker 1",
    "text": "It looks in the end. So it's a before and after, um, or like, it's got this, like, you know,"
  },
  {
    "i": 64,
    "speaker": "Speaker 1",
    "text": "before and after UI, this is a sample clip. Um, astute watchers might realize that this is not"
  },
  {
    "i": 65,
    "speaker": "Speaker 1",
    "text": "the video we uploaded to cloud devs. Um, I was not able to turn around the design in time"
  },
  {
    "i": 66,
    "speaker": "Speaker 1",
    "text": "cause I'm in Tokyo, but, um, you know, this, I, I ended up doing it. So, um, you know, the"
  },
  {
    "i": 67,
    "speaker": "Speaker 1",
    "text": "important thing here is like, this is all just code, right. There's all cloud code that cloud"
  },
  {
    "i": 68,
    "speaker": "Speaker 1",
    "text": "code is writing. So every,"
  },
  {
    "i": 69,
    "speaker": "Speaker 1",
    "text": "there's a transcript for every Jason. Then there's that final edit Jason file that I mentioned,"
  },
  {
    "i": 70,
    "speaker": "Speaker 1",
    "text": "right. Then there's this, um, this is grading these grading files. Uh, then there are a bunch"
  },
  {
    "i": 71,
    "speaker": "Speaker 1",
    "text": "of react components for re motion. There's this final edit that stitches all together. And then"
  },
  {
    "i": 72,
    "speaker": "Speaker 1",
    "text": "it uses NPX re motion render to render the video, apply the color grading, do the re motion, uh, UI"
  },
  {
    "i": 73,
    "speaker": "Speaker 1",
    "text": "and, and put it all together in the end. So, um, yeah, you know, it's, uh, and it, it, it, it, it,"
  },
  {
    "i": 74,
    "speaker": "Speaker 1",
    "text": "it did it. So, um, yeah, that's, that's how it happened. Uh, yeah. Fable is incredible. Um,"
  },
  {
    "i": 75,
    "speaker": "Speaker 1",
    "text": "hoping to, you know, make more video now that it's, it's, uh, it's easier to edit. Let me know"
  },
  {
    "i": 76,
    "speaker": "Speaker 1",
    "text": "if you have any questions. Thanks guys."
  }
]