YouTube2Text

How to Frames: Mastering Start and End Frames in Google Flow (2026) — Transcript

by AI Mind Revolution · 3,740 words · 543 segments · language en · Watch on YouTube

Full transcript

  1. 0:00Of all the different ways you get to
  2. 0:01create AI videos, the start and the end
  3. 0:03frame are the most reliable. Welcome
  4. 0:05back to this channel and on this one we
  5. 0:07will explore the frames tool on Google
  6. 0:09Flow. This is a straightforward direct
  7. 0:12guide. I'm not going to diverge into
  8. 0:13different topics. I'm just going to show
  9. 0:15you what the frame is, how it works with
  10. 0:17[music] examples.
  11. 0:20Everything starts with your frames.
  12. 0:22Think of this as your first most
  13. 0:24important step. If the image quality or
  14. 0:26the idea is poor, well, the video will
  15. 0:29be too. To get the most out of Flow and
  16. 0:32frames, [music] you really need to
  17. 0:33sharpen your image generation skills
  18. 0:36first. We will create a few examples,
  19. 0:38but for the first example, [music] I
  20. 0:39want to create the images right here
  21. 0:41with you. I will go at the bottom. I'm
  22. 0:43using Nano Banana Pro 16:9 and one
  23. 0:45single generation. If you want to use
  24. 0:47Nano Banana 2, this will work as well.
  25. 0:50And [music] this is my prompt. It's just
  26. 0:52a full body shot of a young woman just
  27. 0:54walking on a trail. We have soft natural
  28. 0:56light and subtle shadows and then we
  29. 0:59finish with a natural texture candid
  30. 1:01shot. I just want a lady walking in the
  31. 1:03forest. That's it. If you want the
  32. 1:04prompts and images I'm using for these
  33. 1:06examples, let me know down in the
  34. 1:07comments and I will send you a link. We
  35. 1:09get it back and this is what I get. It's
  36. 1:11just a woman walking in the forest.
  37. 1:13[music] Fine, this will work. Now, then
  38. 1:16you need to create the end frame. This
  39. 1:18will be the starting frame. And to
  40. 1:19create the first frame from the starting
  41. 1:21frame, you have [music] two choices. You
  42. 1:23can use ingredients or go to the edit.
  43. 1:25And I'm going to do both, why not? I'm
  44. 1:27going to go right here, just take this
  45. 1:28image and then drop it as an ingredient.
  46. 1:31Remember we are using Nano Banana.
  47. 1:33[music]
  48. 1:33Since she's not looking at the camera,
  49. 1:35what I want is for her to look at the
  50. 1:37camera and do something like that.
  51. 1:39Simple. This is my prompt, medium close
  52. 1:41shot. The subject is holding both hands
  53. 1:43up towards the viewer. Her expression is
  54. 1:46joyful, smiling, looking directly at the
  55. 1:49lens. And then I provide the instruction
  56. 1:51for the, you know, the arms. Let's run
  57. 1:53the prompt and hopefully we get
  58. 1:54something
  59. 1:55something consistent. While this one is
  60. 1:57going, you have a lot of different
  61. 1:58methods of creating different images by,
  62. 2:01you know, using sheets or using
  63. 2:03references or using ingredients or
  64. 2:05elements. Again, you have a lot of
  65. 2:07different methods. This is not the main
  66. 2:09topic of this video. I've created other
  67. 2:11videos about subject consistency. If you
  68. 2:13want to take a look, go to the cards or
  69. 2:15check my channel. We get it back a
  70. 2:17response and we get something, you know,
  71. 2:19super consistent. It looks like it's
  72. 2:21her, but most importantly, you need to
  73. 2:23take a look at the the background, at
  74. 2:25the setting. On the end frame is
  75. 2:26something super different in the back or
  76. 2:28the subject, you will be getting these
  77. 2:30weird transformations, which in this
  78. 2:32case is something that we don't want.
  79. 2:34So, this is one way of creating the end
  80. 2:36frame from the starting frame. The other
  81. 2:39way is by going directly to the image
  82. 2:41and then running an edit. We don't need
  83. 2:43to add ingredients. I'm just going to
  84. 2:45use the same prompt that we used a
  85. 2:47minute ago. I'm just going to run the
  86. 2:49task. When we get it back, we can see
  87. 2:51it's again still pretty consistent. Now,
  88. 2:53if you ask me which one which method I
  89. 2:55like the most, I like the edit a little
  90. 2:58bit more than creating it with [music]
  91. 3:00ingredients, but as you can see, both
  92. 3:01methods will work. Should you spend a
  93. 3:04lot of time perfecting your start and
  94. 3:06ending frame? 100% yes. Crappy frames
  95. 3:10equal [music] crappy videos. And when
  96. 3:12you are creating the video, no amount of
  97. 3:14prompting will fix those errors. I will
  98. 3:17use this image. If you want to add it to
  99. 3:19your main dashboard, you just need to
  100. 3:20click on save to project and then make
  101. 3:23sure that you select this one before you
  102. 3:25exit. And once back on your dashboard,
  103. 3:27we can see that we have both images. And
  104. 3:30I can say both are, you know, are fine.
  105. 3:33Once you have your frames and you're
  106. 3:35100% sure, your job is to prompt
  107. 3:39whatever is that happens in between. I
  108. 3:41will use this one as my starting. I'm
  109. 3:43going to go at the bottom and select
  110. 3:44video and then use frames 16:9 one
  111. 3:47single output. Now, we have a new model,
  112. 3:49the fast, the light [music] and the
  113. 3:51quality. You know what? I'm going to use
  114. 3:52the light, why not? I'm going to take
  115. 3:54this as my starting frame. I'm going to
  116. 3:56I can just drag and drop to the starting
  117. 3:58frame or I can click on this one and I
  118. 4:01can select it from the menu. Just
  119. 4:03completely up to you. But I like to drag
  120. 4:05and drop. So, this one will be my
  121. 4:07starting and this one will be my ending
  122. 4:09[music] frame. You have an icon right
  123. 4:10here and with this one you can just swap
  124. 4:13the starting for the ending and the
  125. 4:14ending for the starting. And this is an
  126. 4:16important feature. We will talk about
  127. 4:18this in a minute. Of course, we see the
  128. 4:20images and we know exactly what we want.
  129. 4:23Nevertheless, it's a good idea to let
  130. 4:24the model know what we want from [music]
  131. 4:27this sequence. I'm going to go right
  132. 4:28here to the prompt and my prompt will be
  133. 4:31a woman is walking in the forest. The
  134. 4:33camera follows her with this low
  135. 4:35cinematic movement as she walks towards
  136. 4:37the camera. [music] So, we are pretty
  137. 4:39much describing the frames. Let's just
  138. 4:41run the task and see if we get something
  139. 4:43cool. Okay, so let's see.
  140. 4:45She's walking. She needs to look at the
  141. 4:47[music] camera.
  142. 4:48And that's weird, but she needs to open
  143. 4:51her arms. Now, for the most part, it's
  144. 4:54fine. [music]
  145. 4:54The VO it's adding some artistic
  146. 4:57direction, that leaf, but again, for the
  147. 5:00most part [music] it's a good
  148. 5:01generation.
  149. 5:03All right.
  150. 5:05Okay, so as you can see, it's pretty
  151. 5:06simple to work with. You add your
  152. 5:08starting, your ending and then you
  153. 5:10prompt what happens. Things that you
  154. 5:11need to be aware of, the background and
  155. 5:13the subject must match. In this case, we
  156. 5:16have a consistent subject and then the
  157. 5:18background it's consistent as well. If
  158. 5:21the frames are way too different from
  159. 5:23each other, you will be getting wild
  160. 5:25transformations. Unless you want this
  161. 5:27trippy five type of transformations. If
  162. 5:30that's what you want, then go for it.
  163. 5:32When you create a video, it has a
  164. 5:33limited duration. If you prompt for her
  165. 5:35to jump, laugh, spin, clap and then drop
  166. 5:39a line, that will fail or it might fail.
  167. 5:42The model will still try to deliver all
  168. 5:44of that, but it's better if you keep it
  169. 5:46concise. If you have a full body shot at
  170. 5:48the start, a good idea is to stay close
  171. 5:51to the full body shot. You can provide
  172. 5:53completely different shots from the
  173. 5:55starting and the ending, but again, most
  174. 5:58often than not, you will be getting
  175. 6:00somehow inconsistent generations.
  176. 6:02Sometimes you will be getting something
  177. 6:04awesome and sometimes you will be
  178. 6:06getting something, you know, not so
  179. 6:08awesome. And that's just completely
  180. 6:09fine. The thing is that it will cost you
  181. 6:12more credits. So, when you provide
  182. 6:14completely different frames, you can
  183. 6:16really expect some crappy generations.
  184. 6:19Just be aware.
  185. 6:21Okay, now that you know a little bit
  186. 6:23more about frames, let's explore some
  187. 6:25scenarios. The zoom is one of the safest
  188. 6:27motions that you can execute with
  189. 6:29frames. To do this, you need a zoom in
  190. 6:31and zoom out image. This is my example.
  191. 6:34I have a man on a mountain, let's say,
  192. 6:36and then the same man with arms open and
  193. 6:39looking up. You'll notice that the
  194. 6:41person in my images is in a slightly
  195. 6:43different pose. Now, I've created this
  196. 6:45on purpose. If the subject stays in the
  197. 6:48exact same pose, I mean, from a close
  198. 6:50shot and then just a full body shot,
  199. 6:53when you create the video and you zoom
  200. 6:54in or you zoom out, it will be super
  201. 6:57static. It will create a flat lifeless
  202. 6:59video, just scaling the image up and
  203. 7:02down. So, you need to make both shots
  204. 7:05dynamic. You can change a little bit of
  205. 7:06the lighting of what's going on in the
  206. 7:08back. Maybe if you have cars, you can
  207. 7:11add cars in the background or anything
  208. 7:12in the background that is moving. Maybe
  209. 7:14the wind blows and the hair or the
  210. 7:17clothing moves. Let's do it. I'm going
  211. 7:18to take this as a starting frame and the
  212. 7:21other one as an ending frame. And of
  213. 7:23course, we always need to prompt for
  214. 7:25something. A man standing on the peak of
  215. 7:28a mountain and the camera zooms out as
  216. 7:30he opens his arms. Again, we are pretty
  217. 7:33much describing the obvious, but you
  218. 7:37kind of need to do this. It's a way of
  219. 7:38reinforcing the model what we need to
  220. 7:41do. I'm going to still run it on light.
  221. 7:42Let's see what we get back. While this
  222. 7:45one is going, what you can do is to go
  223. 7:47right here at the bottom and you can
  224. 7:48just click and it will swap the frames.
  225. 7:51Now, we have the other motion. Instead
  226. 7:53of zooming out, we will be zooming in.
  227. 7:56I'm going to change the prompt, of
  228. 7:58course, for the camera zooms in as he
  229. 8:00closes his arms. Let's run it. We have
  230. 8:03the first one. Let's just take a look.
  231. 8:07Zoom out. Man opens his arms. It looks
  232. 8:10[music] very natural. And we end on the
  233. 8:14ending frame.
  234. 8:15Cool.
  235. 8:21You can see some issues with the video,
  236. 8:22but remember that we are using the cheap
  237. 8:25version of VO. If you run the quality,
  238. 8:27you will be getting something better.
  239. 8:29Let's take a look at the other one.
  240. 8:30Zoom in.
  241. 8:32It zooms in.
  242. 8:38Well, that was weird. Of course, this is
  243. 8:40a bad generation and the prompt and the
  244. 8:42images are not the problem. It's just
  245. 8:44the way it is. We got a bad generation.
  246. 8:47What can we do? Well, the only thing
  247. 8:48that you can do is to run it again.
  248. 8:51That's it. Let's see if we get it right
  249. 8:53on this generation. And uh that's a
  250. 8:56little bit weird.
  251. 8:58But for the most part, [music] it's
  252. 8:59fine.
  253. 9:02Since we are prompting for camera
  254. 9:03motion, you can and you should be
  255. 9:06specific. By default, if you don't
  256. 9:08prompt for anything, the model will
  257. 9:10create a slow type of movement. If you
  258. 9:12want more energy or maybe a different
  259. 9:13type of movement, you need to prompt for
  260. 9:15it. You can use keywords like dolly,
  261. 9:18pull, arc, crane or fast. I'm going to
  262. 9:21try it right now. I will start with the
  263. 9:23man, you know, the close-up and then the
  264. 9:25other one will end. And my prompt is
  265. 9:28going to be a man standing on the peak
  266. 9:30of a mountain we can know. The camera
  267. 9:32dollies out fast from the subject to
  268. 9:34reveal the mountain. Cool. I'm going to
  269. 9:37run this generation. And while this one
  270. 9:39is going, what if you prompt for a
  271. 9:41different camera motion? What we are
  272. 9:43doing right now is prompting the zoom in
  273. 9:45and the zoom out, which is obviously
  274. 9:47what we have on the first frame and the
  275. 9:48last frame. But what if I go right here
  276. 9:51and do something completely different?
  277. 9:53We still have the man on a mountain, but
  278. 9:54then I'm saying a smooth sweeping drone
  279. 9:57shot that pulls away into the clouds.
  280. 9:59And hopefully you can see the problem
  281. 10:01with this prompt. Because we provided an
  282. 10:03ending frame, the model is forced to
  283. 10:05finish on that very specific shot. If
  284. 10:07[music] you prompt for a completely
  285. 10:09different final scene, it will try to do
  286. 10:12it and then it will be forced to finish
  287. 10:14on the [music] ending frame. Depending
  288. 10:16on the motion that you're prompting,
  289. 10:18this can look great because it will
  290. 10:20create a transition or you will be
  291. 10:22getting something just super weird. It's
  292. 10:24just like throwing a coin. It's like
  293. 10:2650/50. Prompting for something crazy or
  294. 10:28something that maybe they model will not
  295. 10:31understand can give you a good result,
  296. 10:33but most often not it will not. [music]
  297. 10:36So, you need to have the extra credits.
  298. 10:38I'm going to run this generation and
  299. 10:39while this one is going, we can take a
  300. 10:41look at the other one. On this one, the
  301. 10:42camera needs to dolly [music] out fast
  302. 10:45to reveal the background. Let's see what
  303. 10:47we get back.
  304. 10:50It is doing that. We said that we wanted
  305. 10:52to reveal the vast mountain range. And
  306. 10:55it's doing that, but it
  307. 10:56>> [music]
  308. 10:56>> at the same time it's just failing
  309. 10:58because it's just not stopping on the
  310. 11:00ending frame.
  311. 11:01And to [music] finish, because again
  312. 11:03it's needs to be forced to finish on
  313. 11:05that ending frame, it will create a
  314. 11:07transition. And this is the transition.
  315. 11:11Fine.
  316. 11:12Not a good generation, but again not bad
  317. 11:15generation. This is still very usable.
  318. 11:17Now, let's see the other one. So, in
  319. 11:19theory it should be going up into the
  320. 11:21clouds and then at some point go back.
  321. 11:25Let's see.
  322. 11:26And notice it's a little bit weird.
  323. 11:28You're just getting
  324. 11:29>> [music]
  325. 11:29>> clouds.
  326. 11:31And then the transition, of course.
  327. 11:37If we run this generation like 10 times,
  328. 11:40I'm pretty sure that we will be getting
  329. 11:42maybe two that are usable and the rest
  330. 11:45will [music] be unusable. So, when you
  331. 11:47have a very specific starting frame and
  332. 11:49a very specific ending frame, [music]
  333. 11:50trying to prompt for something that
  334. 11:52makes sense. If you provide something
  335. 11:54complex, you will be getting something
  336. 11:56weird and a transition. This idea
  337. 11:58applies to any type of motion. Just to
  338. 12:00give you one more example, I have this
  339. 12:02woman and you have an idea of what I
  340. 12:04want to do. [music]
  341. 12:05I want to take this woman as a starting
  342. 12:07frame and then I want to push in and we
  343. 12:10want to see her eyes with tears. This is
  344. 12:13my prompt. The camera performs a fast
  345. 12:15dolly push in. Keywords fast [music]
  346. 12:18dolly push in towards the eyes of the
  347. 12:21woman. Again, I'm reinforcing what we
  348. 12:23want to do. It's pretty obvious to push
  349. 12:26in towards the eyes, but it's always a
  350. 12:28good idea if you prompt for it. I'm
  351. 12:31going to run this generation. While this
  352. 12:33one is going, we are not providing a lot
  353. 12:34of information of what will happen in
  354. 12:37between. [music] And we are not saying
  355. 12:39what we don't want. So, the AI might
  356. 12:42fill the gaps with some effects or maybe
  357. 12:45some movement from the subject or maybe
  358. 12:48something completely weird. If you want
  359. 12:50for the subject to stay static and the
  360. 12:52background to remain consistent and
  361. 12:54remain the same, you need to go to the
  362. 12:57prompt and say the subject [music] is
  363. 12:59completely static, the background stays
  364. 13:01consistent while the camera pushes in.
  365. 13:03We we get it and this should be a very
  366. 13:05simple generation. But after running a
  367. 13:07lot of different generations, I'm kind
  368. 13:09of a let's say 50% sure that we will be
  369. 13:12getting a weird transition. I'm going to
  370. 13:14go right here.
  371. 13:15We have the motion
  372. 13:17and [music] there we go. That's the
  373. 13:19weird part.
  374. 13:20Nonetheless, for the most part it's it's
  375. 13:23good. It's fine.
  376. 13:26But it's [music] adding a little bit of
  377. 13:27VFX.
  378. 13:29If you don't want for the VFX, you just
  379. 13:31need to say no VFX, no anything, just
  380. 13:34[music] keep the background consistent
  381. 13:35and on and on and on. So, you need to
  382. 13:37prompt for this.
  383. 13:39Fine.
  384. 13:40As you can see in this video, I'm trying
  385. 13:41to show you different examples, but I am
  386. 13:43also expecting to get errors. So, you
  387. 13:46can see where you can find some
  388. 13:48glitches. The next most reliable way is
  389. 13:51a static shot.
  390. 13:53You have an establishing shot, which is
  391. 13:56the where, and then the ending frame is
  392. 13:58the what will happen. As an example, I
  393. 14:00have this scene with an with a couch,
  394. 14:02just an empty couch. And on the next one
  395. 14:05is the complete unchanged scene, but
  396. 14:07then we have a woman sleeping on the
  397. 14:10couch. As before, we can assume what
  398. 14:12will happen. I want an empty couch and
  399. 14:14then I want the woman
  400. 14:16>> [music]
  401. 14:17>> sleeping on the couch. This is my
  402. 14:19prompt. A woman enters the room and
  403. 14:21smoothly lies down on the green couch.
  404. 14:24Her weight sinking into the cushions
  405. 14:26naturally. So, yeah, it's kind of simple
  406. 14:28and this is a resource that we use the
  407. 14:30whole time. We have the shot of
  408. 14:32something and then something happens on
  409. 14:34that something. It doesn't need to be a
  410. 14:36room. It could be an empty car. We have
  411. 14:37a shot of an empty car and then with the
  412. 14:39next frame is going to be someone inside
  413. 14:42the car. In this case, instead of a car,
  414. 14:44we have a room. An empty room, [music]
  415. 14:45someone walks in and performs an action.
  416. 14:48Another thing that you can do, if you
  417. 14:50don't care about the subject and how the
  418. 14:53subject is, because this is part of
  419. 14:55providing an ending frame. We are sure
  420. 14:57that she is sleeping on this very
  421. 15:00specific position, you can, of course,
  422. 15:02just remove the ending frame. And then
  423. 15:04when you prompt for something, the model
  424. 15:06will just fill the gaps for you. And I'm
  425. 15:08going to just run a test. I'm going to
  426. 15:10run this generation and in the meantime,
  427. 15:12we have the other one. Going to play it
  428. 15:13back and I'm 100% sure that this
  429. 15:16generation will be a good generation.
  430. 15:18Going to play it back.
  431. 15:20And it's a very simple [music]
  432. 15:22generation. She needs to lie down
  433. 15:25and
  434. 15:27good vibes.
  435. 15:34Let's go back to the beginning and
  436. 15:36There we go. [music] She goes in. Oh,
  437. 15:39nice couch. I'm going to take a nap.
  438. 15:43And that's it.
  439. 15:45Yeah, it's a pretty safe way of working
  440. 15:47with frames. Let's see the other one. We
  441. 15:50have no idea what is the woman, what
  442. 15:52will happen and where she's going to go.
  443. 15:57Nonetheless, it still is a good
  444. 15:58generation.
  445. 16:00Maybe this part
  446. 16:03it's not and it looks a little bit
  447. 16:05weird. It is just a nice touch.
  448. 16:07It's that she's grabbing the blanket
  449. 16:11from a different from here.
  450. 16:13And of course it fails. That's why
  451. 16:16providing the ending frame will always
  452. 16:18guarantee [music]
  453. 16:20that you get most likely good
  454. 16:23generation.
  455. 16:26Although again, nothing is guaranteed.
  456. 16:28As I mentioned before, it doesn't need
  457. 16:29to be a room or maybe the inside of
  458. 16:31something. It could be a very close shot
  459. 16:34of something else. On this example, I
  460. 16:36have this exhaust and then of course the
  461. 16:38car, you know, we turn on the car the
  462. 16:40car and we see the sparkles and the
  463. 16:42flame. I know that this is a little bit
  464. 16:44unrealistic, but I know [music] nothing
  465. 16:47about cars. I'm going to drop the
  466. 16:48starting frame and I'm going to go to
  467. 16:50the ending frame. And we just need to
  468. 16:52prompt what happens in between. [music]
  469. 16:54Static camera, close-up of a car
  470. 16:56exhaust. I'm going to run the generation
  471. 16:59and see if we get something cool. While
  472. 17:01this one is going, nothing stops you
  473. 17:03from using just the first frame and
  474. 17:06trying to be creative. Just like we did
  475. 17:08with the couch. But again, you can
  476. 17:10expect some glitches. Just to give you
  477. 17:12an example, I'm going to go to this
  478. 17:14woman and I'm going to drop the [music]
  479. 17:16starting frame as the starting frame.
  480. 17:18And I'm going to prompt for something
  481. 17:20way more challenging. The camera
  482. 17:22performs a full 360° orbit around the
  483. 17:25subject, [music] finishing with her
  484. 17:27flashing a big smile and a cool
  485. 17:30energetic pose. So, we are being a
  486. 17:32little bit more creative and let's say
  487. 17:34experimental and try to get something
  488. 17:36super dynamic. This is actually fine,
  489. 17:38but you need to really remember that
  490. 17:40doing things [music] like this will cost
  491. 17:42you a lot of credits. You might get it
  492. 17:44on the first try, but most likely you
  493. 17:47will need to run it multiple times until
  494. 17:49you get what you want. I'm going to run
  495. 17:51this and on the meantime, I will go to
  496. 17:54this one. Let's see if we get something
  497. 17:56cool
  498. 17:57or something weird.
  499. 17:58We get the fire and the sparkles.
  500. 18:02That was a little bit weird, but it's
  501. 18:04fine. It's usable.
  502. 18:06We don't need to use the whole clip.
  503. 18:08Sometimes you just need parts.
  504. 18:10Maybe we can just stop it right here and
  505. 18:11then move on to a different shot.
  506. 18:15And then we get the sparks, of course.
  507. 18:17Usually, almost everything that you get
  508. 18:19it's usable. You just need to take the
  509. 18:21different [music] parts and then go to
  510. 18:23the editor and see if you can create
  511. 18:25something cool. If you can't, doesn't
  512. 18:26matter. You can go back, create a new
  513. 18:28generation until you get something that
  514. 18:30you think it will work and then take it
  515. 18:32back to the editor. Let's see the 360.
  516. 18:34And again, this is just like flipping a
  517. 18:36coin.
  518. 18:40>> [music]
  519. 18:40>> Okay.
  520. 18:42It was actually pretty cool. We still
  521. 18:45got to get that
  522. 18:46effect.
  523. 18:47>> [music]
  524. 18:51>> It wasn't that bad. And this was the
  525. 18:53first generation. So, sometimes you will
  526. 18:55be getting on the first and sometimes
  527. 18:57you will need to run it multiple times
  528. 18:59until you get it. That's the way it
  529. 19:01works. Right, so this was an
  530. 19:03introduction to frames and flow. But
  531. 19:05this applies to any other platform.
  532. 19:07There's so much we can do right here
  533. 19:09like 2D animations, looping, time lapses
  534. 19:13and a lot more. If you want for me to
  535. 19:15upload videos showing you different
  536. 19:16examples that we can create with frames,
  537. 19:18let me know down in the comments.
  538. 19:20[music] All right, so if you liked all
  539. 19:21of this and you learned something new,
  540. 19:23remember to like, subscribe and leave a
  541. 19:25comment. Don't be a stranger.
  542. 19:35>> [music]
  543. 19:41[music]

About this transcript

This page contains the full transcript of How to Frames: Mastering Start and End Frames in Google Flow (2026) by AI Mind Revolution, generated from the public captions YouTube serves with the video. The transcript has 3,740 words across 543 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.