Today we are talking about the "full banquet" in the field of "video AI" - Google Veo3!
This tutorial goes through the registration of Veo3, from basic use to advanced play, and packs in 12 sets of "famous scene" cues.
I'm going to share an original JSON format English cue word with one click changed to Chinese cue word, not only play Veo, but also can be easily converted to Wan, that is, dream use version.
I. Getting Started with Veo3
Google, for the past two decades, the world's dominant Internet company.
However, in just two years, when the hurricane of generative AI swept in, Google once seemed to stagger, and was even quietly mocked by many people, Google is old!
Recently, however, Google has gradually found its feet, and with the Gemin2.5 pro and Veo3 two big killers, it has gradually taken the initiative to compete, forcing OpenAI to release the GPT5 in the near future, as soon as next week.
I. Introduction to Veo3
In April 2025, Google launches Veo3, the first native "audio-video synchronization" video model.
What is audio-visual synchronization?
Before Veo3, AI video is the "silent era", can only do video, sound can only rely on post-production editing, without using a very clumsy way to achieve lip sync, the effect is very collapse.
Veo3 can generate an 8-second movie-quality video + character dialog (supports multiple people) + sound effects directly from the text prompts.
And again, Veo3's great power.
There are two main schools of thought in the jungle of AI doing video:
- Cultivate "Tusheng Video" - make photos move. The quality of the finished movie is limited by the quality of the picture and its spatial expression
- Famous "Vincent Video" - "shooting a movie out of thin air", high degree of freedom but difficult to control the details, the quality of the early versions was a complete disaster.
Veo3 is a double whammy: it can do both text-to video and graphic-to video. In particular, text to video + sound is a killer feature of Veo!
Bold prediction, Veo3 will be an iconic product!
Prior to Veo3, graphic video was the mainstream, and since Weo3, text-based video will gradually become the mainstream!
II. Veo3 Registration
To use Veo3, you must be a paid member of Google, $19.90 per month
I've researched the white whoring spree for you as follows:
1. Small white out-of-the-box ready-to-eat version:
Go to the idle fish search key words "Gemini Pro", spend a cup of milk tea money to buy a finished number, white whoring 15 months, angry save $300!
The finished product has been taken care of: Google account creation, Google email activation, educational email verification, and 15-month trial application all in one, and you can use it right away.
2. Technical Bull Unlimited White Whore Edition
First level, the heart of the path is formed, 3 times a day for Veo3.
The second step is to go from the path to the devil and whore out Veo3 50 times a month through the Google flow platform.
The third level, from the devil to the Tao, jumps across the board via Google Drive, documents, multiple Gmail mailboxes, and unlimited use of the
Where to play Veo3?
Two websites: Gemini and Flow
- The Gemini website is Google's flagship AI platform, with features that include conversations, a text editor, and deep research,
Image model Imagen, video model Veo3
- Flow Website It's a video editing site from Google. 1000 points/month, you can do more than one, but when it runs out, it's gone
After completing the registration, the user can use Veo3 on both of the above mentioned websites at the same time.
Gemini is suitable for most of the students as it can do videos, pictures, conversations, and research papers.
If you're a heavy user of video AI, consider looking into using Flow
Veo also comes in versions, Fast and Standard.
On the Gemini platform, you can only do the Fast version, and only 3 videos per day per pro account.Flow
The platform can be done in Fast or Standard version, 1000 points/month while supplies last.
1. Gemini website
https://gemini.google.com
After logging in, the main interface is as follows:
Click on "Videos with Veo" to enter the video making console.
Vincent Video
Chinese prompts are sufficient for Gemin, while English prompts are required for Flow.
As shown in the picture below, I type in Chinese and it takes me 2-3 minutes to generate an 8-second video.
Note that Veo3 now only supports 8 seconds of video in 16:9.
Figure video
Veo3 also supports graphic video, upload a picture of a beautiful woman and enter a Chinese prompt word:
"Drone footage, fast traversing, from the back of the woman over her head, down the fishing pole into the surface of the sea water, where a shark is biting and struggling with the bait"
In a few minutes, the video was ready, video, shipping mirror, sound effects all at once. Especially the sound effects, the sound of waves, fishing rod winch, sharks struggling, immersive!
2、Flow official website
Website: https://labs.google/fx/tools/flow
Select mode on the left
Supports born-text video, born-image video, first and last frames; $200/month version also supports multi-image referencing
Select model on the right.
You can choose Veo2, Veo3, and unlike the gemini site where you can only choose the Fast version, here at Flow you can choose the standard version of Veo3 for 100 points each! One Veo3 is worth the cost of 5 Fast versions!
The basic operations of Flow are similar to those of gemin, but we'll talk about the more advanced operations some other time.
IV. Cue words for Veo3
Cues are used to describe the content and movement contained in the video. Veo3 understands very detailed cues, and the more complete, precise and informative the description, the higher the quality of the generated video and the closer it is to the desired content.
Regarding AI authoring (scenarios like text-to-graph, graphic-to-video, etc.) inCue ConstructionThe system of formulas is divided into three categories:
- basic formula: In its simplest form, cue word = subject + scene + motion. For example, if you want the AI to generate "cat (subject) jumping (movement) in the garden (scene)", you can combine the base commands like this.
- advanced formula: For more detail and to enhance the texture of your work, cue words = subject (subject description) + scene (scene description) + movement (movement description) + aesthetic control + stylization. For example, "Cat with fluffy fur (subject + description) bouncing lightly in a garden full of flowers and dappled sunlight (scene + description) (motion + description), warm light and shadow (aesthetic control), oil painting style (stylization)" allows AI to produce content with a more graphic and unique style.
- Toussaint Video FormulaFor the "Generate Video from Pictures" scenario, cue word = Motion + Lens shift. For example, if you have a still picture of a cat and you want to turn it into a video, you can command "cat walks slowly (motion), the camera pulls away from close-up to full body (lens movement)" to control the video dynamics and lens changes.
1、Chinese support
Veo3 supports Chinese prompts on the gemini platform.
The Flow platform is English only and does not support Chinese.
2. The more detailed the cue word, the better
A format called JSON is popular in foreign Veo3 circles.
A standard JSon format looks like this, and a lot of white people look at it like this.
3、JSON format into Chinese
Such a prompt word has several shortcomings, white people are not convenient to understand, not good to modify, the domestic dream, can only support 500 words, similar to the Json prompt word is too long, can not be used to the domestic platform!
I originated a modified cue word:
One click to change Veo English Json to Chinese
Cue Words: Are you a professional cue word optimizer, please convert the following from JSON format to plain natural language cue words for mapping or video generation. The requirements are as follows:
-Remove all JSON structures (including key names, quotation marks, curly brackets, colons, etc.) and retain only the core elements of the content; and rewrite them uniformly as coherent, concise, natural language prompts with a sense of picture;
-If the original text exceeds 500 words, please compress and streamline it to 500 words or less, trying to retain key details, style and theme;
-Remove content that is redundant, repetitive, too technical, or unrecognizable by AI. Retain keywords that are appropriate for AI image or video generation, such as character, scene, mood, style, and action.
Please optimize the following JSON prompt words according to the above rules and output the results.
You can open kimi or beanbag, with the above tips, the Json version to the Chinese version of the compressed version, such as the previous paragraph, kimi to you directly to the Chinese version of the reduced version, as follows
02:30 a.m. Zoo surveillance gray-green night vision. Wide angle overhead shot of the enclosure: a rhino strides slowly across the slippery ground, a Humboldt penguin standing steady on its dorsal fin. Drizzle drips on the lens, water droplets slide along the night vision spot. The rhino steps into a puddle of water, causing dark silver ripples, and the penguin balances on its flippers. at 5 seconds the rhino grazes the stand, causing the frame to flicker and the scan line to distort, recovering a second later. The rhinoceros stops and sprays cold mist; the penguin probes the camera, and its little hop remains in place. No music, just the sound of rain, low noise and distant fan hum."
Input Veo3, get the following video, although the effect is slightly worse than the English JSON prompt word, but the advantage is easy to understand, can be copied to the domestic platform use
4. Continuous plot cue words
If you want to make a theme out of a character, you can use kimi or deepseek to generate continuous prompts in the following format
Tips:
You are asked to play the role of a top VLOG director who is imaginative and well versed in the language of the movie. Help me write a detailed English cue for Veo3 in about 8 seconds. The video is shot from the handheld lens of bigfoot's selfie stick, the subject is bigfoot's daily fish catching vlog from the selfie forest, it is a selfie style video capturing bigfoot's funny moments of the day, keeping the tone lighthearted, casual and cinematic, and it needs to have 5 seconds of bigfoot's lines that he narrates to the camera!
5、Picture comes with tips
A few days ago, netizens discovered another feature of Veo3 Cowhide. As long as the text is well labeled on the picture, after Veo gets the picture, it will make a video according to the text annotation of the picture, and then erase the text in the video.
Prompt:
High-intensity action scene. Motion 1: camera pulling back slightly as Car speeds towards us. Motion 2: Camera pushes in and catches Car drifting hard as it's rear whips around around a hairpin turn. Motion 2: Camera pushes in and catches Car drifting hard as it's rear whips around around a hairpin turn. Motion 3: The camera chases the car around the turn from low side angle. " Immediately delete instructions in red on the first frame and execute in order"
Veo generates perfect videos:
II. Veo3 Classic Cue Words
Because it's a Vincentian video, and because it's clever, Veo is very playable.
From the beginning of the selfie Vlog, cut fruit, to the recent explosion of products, text, global netizens constantly discover a variety of new ways to play!
Today we introduce you to 12 classic Veo3 ways to play, with hints and examples!
1. POV Direct Broadcast
Three Kingdoms Frontier Interview Prompts
prompt: "First-person perspective, a Chinese girl in a windbreaker holding a selfie stick on a warship at the Battle of Red Cliffs, filming herself surrounded by raging flames and collapsing soldiers. She extends the selfie stick to show her shocked face, then rotates to
She says in a tense tone, "
This battle is total chaos!" The background features sounds of clashing swords and burning flames, presenting a cinematic historical The background features sounds of clashing swords and burning flames, presenting a cinematic historical style."
Chinese prompt: "First person view of a Chinese girl in a rushed suit holding a selfie stick on a warship at the Battle of Red Cliffs, filming herself, surrounded by roaring flames and crumbling soldiers. The hand extends the selfie stick to show the girl's shocked face, then rotates to capture the burning wooden mast and the confused shouts of the soldiers. In a nervous tone, the girl says, "This battle is chaos!" Sound of swords clashing and flames burning in the background for a cinematic, historical style"
Bigfoot Vlog Cue Words
Selfie-style vlog film shot on Bigfoot's handheld selfie stick-a playful, cinematic glimpse into his daily life in the forest. Close-up POV as Bigfoot wades through a sun-dappled river, his furry arm occasionally entering frame. Suddenly, he lunges, splashing water, then lifts a wriggling fish triumphantly. He grins at the camera (breathing heavily, wet fur glistening) and says in a deep, gravelly voice: 'Breakfast is served. 'Breakfast is served... nature's fast food!' Keep the tone casual, documentary-style with warm, natural light, subtle camera shakes. Keep the tone casual, documentary-style with warm natural light, subtle camera shakes, and immersive sounds (rustling leaves, flowing water). End with a messy bite of the fish-raw and wild
2. Cutting fruit
Fruit Cutting Cues
Prompt: "Shot in extreme macro perspective, a flawless, crystal-clear, and detail-rich [glass strawberry] rests on a wooden cutting The camera gently pushes and pulls, capturing the rainbow-like glimmers and prismatic refractions on the surface of the [[insert object, e.g., glass strawberry]. insert object, e.g., glass strawberry]. The angle alternates between top-down and a 45-degree side view, freezing the moment the knife slices into the fragile "[glass strawberry]," revealing its intricate internalization. ," revealing its intricate internal textures.
Each slow, deliberate cut produces a crisp "ting"-the sound of glass shattering-followed by the soft clinking of tiny shards and the deep, resonant scrape of steel against wood. Each slow, deliberate cut produces a crisp "ting"-the sound of glass shattering-followed by the soft clinking of tiny shards and the deep, resonant scrape of steel against wood. These sounds layer together to create a rich ASMR soundscape "
Cue word: "Shot from an extreme macro perspective, a flawless, crystalline and richly detailed [glass strawberry] sits atop a wooden cutting board enveloped in warm light. The camera slowly pushes and pulls to capture the iridescent shimmer and prismatic refraction of the [glass strawberry] surface. The camera angle switches between a top view and a 45-degree side view, capturing the moment when the blade cuts into the fragile [Glass Strawberry], revealing its delicate internal texture.
Each slow, deliberate cut is accompanied by a crisp "ting" - the sound of breaking glass - followed by the slight clash of tiny shards and the deep, reverberating scrape of the steel blade against the wood. These sounds are layered on top of each other to create a rich and detailed ASMR soundscape. "
3. Girlfriend ASMR Whisper
Girlfriend Whispering ASMR Cue Words
Cue word: American country house fireplace burning, on the couch, a Chinese 20 year old kpop gri in spaghetti strap lace pajamas lying on the couch, shot from her handheld angle, slowly getting up while saying in a soft (ASMR style) voice: "bad boy, come on!
4. Terror Yacht
A beautiful woman takes a selfie and is suddenly attacked by a great white shark cue word
Cue word: "POV selfie shot of a young kpop girl on a speedboat, smiling into her phone: " very very good" Clear blue ocean, bright sun. Ocean, bright sun. Suddenly, a massive great white shark breaches silently behind him, jaws wide open, inches from her head. Sound: wind, engine, excited shout, splash, deep shark growl.Style: vlog realism, high tension, comedic terror. no subtitles"
5. World Slapping Contest
Slapping World Series Prompts
Scene Description.
Ultraman stands calmly at the podium. It lifts one long, heavy arm, pauses for a moment, then swings hard across
The slap lands flush. The man's neck twists violently and he drops without resistance.
Visual Style: real broadcast footage.
Camera Movement: locked mid-shot, slight jolt at the moment of impact.
Main Subject.
Ultraman palm landing clean across the man's face with full mass behind it.
Background Setting: professional slap contest stage, spectators pressed forward in anticipation.
Lighting Mood: intense, high-contrast lighting over both contenders.
Audio Cue: deep crowd hum, then a heavy THWACK that silences the room - followed by stunned crowd noise."
6. Pixel people dancing together
Cue word for framing with pixelated people
{ "concept": "Inside a Korean supermarket at night, a 23-year-old Korean woman and a pixelated 16-bit Yeti perform a trending TikTok dance in the middle of the aisle. Shoppers stop and watch in awe, amusement, and disbelief as the real and digital worlds collide. ", "shot": { "composition": "medium-wide frontal shot framed down the aisle, centered on the dancers, depth created by store shelves and scattered onlookers", "camera_motion" ": "slow lateral dolly or static with slight handheld feel to mimic organic bystander filming", "frame_rate ": "30fps", "film_grain": "minimal, clean digital look with soft texture"}, "subject": { "character_1": { "description ": "23-year-old Korean woman with black ponytail, wearing a bright cropped hoodie and black leggings with white sneakers, energetic and confident" } energetic and confident" }, "Character_2": { "description": "16- bit pixelated Yeti, chunky and expressive, white fur blocks, cyan glowing eyes, perfectly mirroring her choreography with a cartoon twist " } }
"scene": { "location": "well-lit Korean supermarket, night time, convenience- style layout with narrow aisles and fluorescent lighting", "time_of_day": "night ", "environment": "brightly stocked shelves with snacks, noodles, and drinks. Neon sale signs, open refrigerators buzzing softly, and a large, clean, and clean room. Neon sale signs, open refrigerators buzzing softly, and a few scattered baskets and carts" }, "visual_details": { "action": "The woman and pixel Yeti perform a precise, upbeat TikTok routine in sync. Around them, shoppers pause to watch-some filming. Around them, shoppers pause to watch-some filming on phones, others reacting with surprise or laughter. One child tries to copy the moves. pixelated flourish at the end, tossing pixel confetti", "props": "shopping carts, stacked The Yeti adds a final pixelated flourish at the end, tossing pixel confetti", "props": "shopping carts, stacked ramen boxes, sale banners, refrigerated drinks glowing with light, smartphones raised by onlookers" }, "cinematography ": {
"lighting": "fluorescent top-down supermarket lights, flat and clean, with some natural bounce from packaging and product labels", "tone": "fun, energetic, surreal slice-of-life with digital packaging and product labels", "tone": "fun, energetic, surreal slice-of-life with digital twist" }, "audio": { "music": "upbeat viral TikTok track with crisp twist" }, "audio": { "music": "upbeat viral TikTok track with crisp beat drops", "sfx": "grocery cart wheels, phone camera shutters, subtle refrigerator hum, faint crowd murmurs and laughter". faint crowd murmurs and laughter, pixel sparkles with each Yeti move"}, "color_palette": " bright commercial tones-white light, colorful product packaging, primary clothing colors, soft neon accents from packaging and coolers "}
7. Miss Dance and Sing
Street Dance Cue
{ "shot": { "composition": "Medium tracking shot, 50mm lens, shot on RED V-Raptor 8K with Netflix-approved HDR setup, shallow depth of field", "camera_motion": "smooth Steadicam walk-along, slight handheld bounce for naturalistic rhythm", "frame_rate": "24fps ", "film_grain": "clean digital with film-emulated LUT for warmth and vibrancy"} , "subject": { "description": "A striking young woman with flawless dewy skin and a Her long black hair is styled in a side-parted soft wave, falling over one shoulder. She has large monolid eyes framed by delicate winged eyeliner and pearlescent shimmer, glossy gradient lips in cherry rose, and glass-like skin. She performs with idol-like precision - expressive facial control, hand gestures, and a sleek V-line jaw. expressive facial control, hand gestures timed to rhythm, eyes occasionally darting to camera.", "wardrobe" : "Pearl-trimmed white crochet halter top, structured high-waisted denim shorts with a slim studded tan belt and red lacquered star buckle , oversized red gingham shirt slipped off the shoulder, sheer nude tights, white chunky platform sneakers, silver chain earring on left ear, and metallic rose gold nail polish." }, "scene": { "location": "a quiet urban "environment": "empty sidewalks, golden sunlight reflecting off puddles and windows, occasional birds fluttering by, street slightly wet. fluttering by, street slightly wet from overnight rain"}, "visual_details": { "action ": "she sings directly to camera with confident rhythm, her mouth clearly articulating Korean lyrics in perfect sync, smiling and gesturing in time with the smiling and gesturing in time with the beat, walking with bounce in her step", "props": "morning mist, traffic light turning green in the mist, traffic light turning green in the distance, reflective puddles, subtle sun flare" }, "cinematography". : { "lighting": "natural golden-hour lighting with soft HDR bounce, gentle lens flare through morning haze ", "tone": "playful, stylish, vibrant", "notes". "STRICTLY NO on-screen subtitles, lyrics, captions, or text overlays. Final render must clearly show active vocal performance, not Final render must clearly show active vocal performance, not background track mimicry." }, "audio": { "ambient": "city birds chirping, distant traffic hum, her boots tapping pavement", "voice": { "tone". "bright, teasing, confident", "style": "K-pop vocal phrasing with flirtatious intonation, crisp Korean diction, playful swagger" }, "lyrics": "I'm not NJ Red, I'm a beautiful girl from Korea, let's have fun together! " }, "color_palette": "sun-warmed pastels with vibrant reds and denim blues, soft contrast with warm film LUT ", "dialogue": { "character": "Woman (singing in Mandarin) ", "line": "I'm not NJ Red, I'm a beautiful woman from Korea, let's have fun together!" , "subtitles": false }, "visual_rules": { "prohibited_elements". [ "subtitles", "captions", "karaoke-style lyrics". "text overlays", "lower thirds", "any written language appearing on screen "] }}
8. Gift advertising
Product Full House Prompt Words
Main Description: The scene begins in a sunny, empty room. In the center sits a cardboard box with a cute young girl's face drawn on it with a marker. In the center sits a cardboard box with a cute young girl's face drawn on it with a marker. The box wiggles slightly, lets out a soft girlish giggle, and then bursts open! Countless colorful hair ties, girly accessories, soft pillows, plush toys, and vibrant trinkets gush out. A dreamy vanity "grows" rapidly from the floor like a real tree, its mirror gleaming A dreamy vanity "grows" rapidly from the floor like a real tree, its mirror gleaming softly; a pink-and-white bookshelf assembles itself and stands in a corner; ribbons and fairy lights dance through the air before landing in a storage basket. Once everything is perfectly in place, a casually dressed young girl in an oversized hoodie and pleated skirt walks in with light steps, curiously observing her new paradise, and her new life. observing her new paradise, and finally curls up comfortably on a cloud-soft beanbag.
Core Elements: Cardboard box with a girl's face drawn on it, dreamy vanity, pink-white bookshelf, various soft pillows and beanbags, wall- mounted girly shelves, a window hammock chair, a motion-sensing night-light, girly accessories (hair ties, charm bracelets, plush stars, a heart- mounted girly shelves, a window hammock chair, a motion-sensing night-light, girly accessories (hair ties, charm bracelets, plush stars), a heart-print rug. print rug.
Motion & Atmosphere: Full of playful fun and whimsy, various girly supplies assemble automatically in a lively, charming way, ultimately creating a warm, cozy, and safe paradise built perfectly for a young girl. Motion & Atmosphere: Full of playful fun and whimsy various girly supplies assemble automatically in a lively, charming way, ultimately creating a warm, cozy, and safe paradise built perfectly for a young girl.
Keywords: 16:9, one take, no cuts, dynamic assembly, photorealistic, girl, cute, girl's room, dreamy paradise, heartwarming, no text.
9. Flying machines
Future Flyer Cue Words
A hyperrealistic, 8-second cinematic shot of a squadron of futuristic hoverbikes racing across a vast, windswept desert canyon on a terraformed Mars.", "atmosphere": "A dry, dusty environment with low gravity. ", "atmosphere": "A dry, dusty environment with low gravity. Distant mesas and canyon walls rise Distant mesas and canyon walls rise in the hazy sunlight."}, "scene_details": {"setting": "A wide shot Giant rock formations and mesas rise dramatically in the background, partially obscured by dust kicked up by the hoverbikes. The terrain is uneven, with loose rocks and patches of reddish sand.", "mood": "Fast-paced and exhilarating, the terrain is not a good one. The terrain is uneven, with loose rocks and patches of reddish sand.", "mood": "Fast-paced and exhilarating, conveying a sense of speed, adventure, and the vastness of an alien landscape.", "action_ breakdown": {"timeline breakdown": {"timeline": [{"time": "0-2s", "action": "The lead hover". "action": "The lead hoverbike, closely resembling the one in the image, speeds into the frame from the left, flying low over the terrain and kicking up the hill. The camera begins to track alongside it."}, {"time ": "2-5s", "action": "The lead bike navigates a small rise in the terrain, briefly becoming airborne. The lead bike navigates a small rise in the terrain, briefly becoming airborne with its anti-gravity systems flaring with a subtle blueish glow. Other hoverbikes in the squadron follow closely behind, creating dynamic dust trails and weaving around obstacles."}, {"time": "5-8s ", "action": "The squadron accelerates, their sleek, futuristic designs cutting through the Martian air. The camera angle shifts slightly to show more of the vast canyon landscape as the bikes speed towards the horizon, their red and blue running lights visible through the dust. lights visible through the dust."}]}, "vehicle_and_rider_details": {"hoverbike_design ": "Aerodynamic, low-slung hoverbikes with visible anti-gravity emitters and intricate mechanical details, consistent with the provided image. They have a mix of metallic gray, black, and red accents.", "rider_outfits": "The riders are clad in black, black, and red. "The riders are clad in advanced, dark grey and black armored suits with enclosed helmets, suggesting an exploration or reconnaissance team. Their postures are leaned forward, indicating high speed.", "camera_and_shot": {" camera": "Virtual camera mounted on a fast-moving drone or chase vehicle.", "shot_type ": "Dynamic tracking shot with slight panning movements to keep the lead bike in focus while showing the squadron and environment.", "movement": "Smooth, fast-paced camera movement that follows the action, conveying the speed of the hoverbike. Smooth, fast-paced camera movement that follows the action, conveying the speed of the hoverbikes. Subtle bumps and vibrations can be added for realism.", "focus": "Sharp focus on the lead bike. "Sharp focus on the lead hoverbike, with a slight motion blur effect on the background to emphasize speed."}, " sound_design": {"vehicle_sounds": "A futuristic blend of high-pitched whirring from the anti A futuristic blend of high-pitched whirring from the anti-gravity generators and the low rumble of powerful internal engines. accelerate."
10. The streets of Tokyo
Tokyo Street Words
Cue word: Continuous single take, camera tilting, rolling, flying through a narrow alley in Shinjuku, Tokyo. Neon kanji signs flicker overhead, steam rises from ramen stalls, chefs slice pork and toss noodles into boiling broth. Neon kanji signs flicker overhead, steam rises from ramen stalls, chefs slice pork and toss noodles into boiling broth. camera glides past diners slurping, dives through a paper lantern into a karaoke Camera glides past diners slurping, dives through a paper lantern into a karaoke-inside, someone sings J-pop under blue strobe lights. Camera somersaults through the mic into an underground arcade, claw machines, flashing, retro game sounds blipping, teens cheering. No cuts, impossible camera moves, seamless transitions, energetic and cinematic. 8K high definition definition
11. Rampage
Street Racer Cue Words
Low-angle tracking perspective, fast-paced dynamic follow shot, a vintage sports car in gleaming crimson races through tight city streets, weaving The car's polished body reflects bursts of city light and neon signs as it accelerates under overpasses. The car's polished body reflects bursts of city light and neon signs as it accelerates under overpasses and through narrow alleys. Audio features the aggressive roar of the high-revving engine, gear shifts snapping cleanly, tires screeching, and the rush of wind as it blazes past. Hard, contrasty lighting with sharp highlights and cinematic motion blur. Cinematic style: high-energy chase sequence with retro flair, reminiscent of Ronin or The Italian Job.
12. Sprouting
Sprouting Cue Words
{ "scene": "macro close-up of rich soil", "timeline": [ { "time": "0s", "description": "Fresh, dark soil with subtle surface texture. Tiny green sprouts begin emerging slowly from the ground. Lighting is soft morning light, golden and natural."}, { { "time": "0s", "description": "Fresh, dark soil with subtle surface texture. "time": "2s", "description": "Sprouts grow taller and begin developing small leaves. The soil shifts slightly as roots begin to form underground. Time of day brightens slightly to late morning." }, { "time": "2s", "description": "Sprouts grow taller and begin developing small leaves. " }, { "time": "4s", "description": " Leaves expand into bushier carrot tops, swaying gently. Small orange bulges (carrot roots) become subtly visible above the soil line. Lighting is now midday with stronger highlights." },{ "time": "6s", "description". ": "Fully mature carrot greens above ground. Carrot tops are thick, vibrant, and healthy. A hint of condensation on leaves. Soil shows signs of moisture and root pressure." }, { "time": "8s", " description": "Timelapse ends with a fully grown cluster of carrots in the soil. Orange roots partially visible at the base. Lighting softens slightly as time shifts toward golden afternoon." }], "style": {"lens ": "macro", "lighting": "natural, soft timelapse from morning to afternoon", "depth_of_field": "shallow", "texture ": "ultra-detailed soil and plant structure", "motion": "smooth, organic time-lapse growth" }, "texture": "texture": "ultra-detailed soil and plant structure", "motion": "smooth, organic organic time-lapse growth" }, "duration": "8s"
}
13. Nightclub celebrities
Nightclub celebrity cue word
The camera follows the young model from the rear at 45°, dressed in a jeweled pink silk dress, long straight black hair and crystal earrings, walking quickly along a matte black velvet corridor towards a mirrored elevator to the left, with a flourish LED strip on the floor reflecting the glitter of the jewels. \
The camera stops at the 2-second mark, she presses open the elevator, the soft D5 bell rings, and inside the door five bare-chested Korean male models greet her with smiles. \
4 seconds cut view inside the elevator, the female model half-steps into the door, playfully looking at each other and then suddenly turning around. \
6 seconds of low-angle follow-up shots of her lifting her skirt and running wildly, with her long skirt and earrings flying; the elevator door closes in 7.2 seconds, and the male models look at each other with a smile. \
The scene is warm pink 3300 K, satin flutters in real time, skin and gemstone highlights sparkle; 72 BPM ethereal strings with heartbeat taiko drums, footsteps and metal doors echo in the hallway.
14. Cozy Family
Warm Family Reminders
Continuous single take, the camera glides through the open front door, skimming across the living room first-a child curled on the rug, eyes fixed on the flickering black-and-white cartoons on an old TV. The lens eases right, slips through the doorway into the kitchen; the wife stands with her back to us, slowly stirring the bubbling white porridge, steam blooming softly. The lens eases right, slips through the doorway into the kitchen; the wife stands with her back to us, slowly stirring the bubbling white porridge, steam blooming soft warm light over the retro tiles. The move keeps forward to the fridge in the corner-camera follows me as I step to the aged refrigerator, pull the door; a puff of cold air meets me, I lift out one beer, beads of I drink and drift onto the balcony, looking out the window.
Visual style: VHS grain, pastel hues, gentle analog shimmer.
Sounds: porridge simmering, cartoon chatter from the TV, fridge hum, clink of bottle against teeth.
Mood: nostalgic, quiet, everyday yet vividly alive.
Chinese:
In one continuous shot, the camera slides through the open front door, skimming first through the living room - a child curled up on the carpet, his eyes fixed on the flickering black-and-white cartoons on the old TV. The camera pans gently right, through the doorway into the kitchen; the wife's back is to us, slowly stirring the gurgling white porridge, the steam blooming softly on the vintage tiles. Continuing to pan forward to the refrigerator in the corner - the camera follows me as I walk up to this old fridge and pull open the door; a blast of cold air hits my face as I lift out a bottle of beer, water beads sliding down the bottle. I sip my drink and wander out onto the balcony and look out the window.
Picture style: VHS grain, pastel wax tones, soft analog microflashes.
Sounds: porridge gurgling, TV cartoon dialog, fridge buzzing, bottles clinking against teeth.
Vibe: Nostalgic, quiet, everyday yet vibrant and lively.
This is just the tip of the iceberg of Veo, more high-powered gameplay is on the way, let's talk about it some other time.