Casi listo para ir
Cargando...

The Short Film That Needed No Set, No Camera, No Crew. One Day, with Zoviz

The Short Film That Needed No Set, No Camera, No Crew. One Day, with Zoviz

The film is called 06:00. A young man wakes to his alarm, dresses for work, and steps into a city where every other human being has vanished. Over three minutes his fear melts into freedom — an empty police car, a grand café all to himself, a luxury convertible — until something falls out of the sky and ends Episode One on a hard cut to black. Every frame was generated in Zoviz Studio. This article is the complete recipe.  

Story by Masoud Hatami — https://www.imdb.com/name/nm4699717/

Produced in Zoviz Studio (Seedance 2.0) — https://studio.zoviz.com

The eight rules of cinematic AI footage

Before writing a single prompt, we distilled current AI filmmaking research into eight rules. They are the difference between footage that reads as photographed and footage that reads as generated.

1 · Physics, not feelings. Never write "he is afraid." Write "his shoulders tense, his jaw tightens." The model renders what it can simulate.
2 · One camera move per shot, described by rhythm — slow, smooth, gradual — never by gear specs. Keep camera movement separate from subject movement.
3 · Light is physical. Direction, temperature, falloff: "cold blue ambient, one warm tungsten pool, deep shadow elsewhere."
4 · Timeline prompting. Timestamped beats — [0s], [5s], [11s] — one camera position and one action per beat.
5 · Film language anchors: "ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights." Words that degrade output: cinematic alone, epic, beautiful, 8K, masterpiece.
6 · Positive constraints. "Maintain consistent facial features throughout" — never "no face drift."
7 · Sound is written into the shot, listed at its timestamps. An empty city is an audio effect as much as a visual one.
8 · Storyboard with stills first. Lock faces, locations, and grade as images before spending a second of video.

Step 1 · Break the script into 15 second shots

Seedance 2.0 in Zoviz Pro Studio generates 4 to 15 seconds per run — so a short film is a stack of shots, exactly like a real one. We compressed the script into fourteen shots of 10 to 15 seconds, each carrying one dramatic beat: the alarm, the departure, the reveal, the bus stop, the growing silence, the bicycle, the dead city center, the unanswered phone, the police station, the siren, the joyride, the café dream, the idea, and the fall. Total: about 3 minutes 15 seconds — a real episode length for social platforms.

The craft rule: each shot can hold 3 to 4 internal beats (Seedance cuts within a clip when the timeline prompt asks it to), so fourteen generations deliver roughly forty distinct story moments.

Emre, our lead — an actor who does not exist — realizing the street is empty. Straight from the finished film.

Step 2 · Write the film bible

One page, decided before anything is generated. The character: Emre, 20 — short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, identical in every shot. He never speaks; all acting is physical. The format: 16:9, 480p drafts, 24fps film feel. The palette: steel blue shadows, concrete gray, one warm amber accent that follows him through the film. The arc in light: cold and empty at the start, warming as fear becomes freedom, cut to black at the end.

Step 3 · Cast the actor and build the world

Ten images in the Quick Image Generator (High quality — same credit price, better output) become the film's entire art department. Three of the actor, one style frame, six locations. The trick that locks the actor's face: generate the first portrait, then attach it as a reference image while generating the other two angles.

Casting and set design as still images: the actor sheet and the bedroom with its 6:00 alarm clock

R1 · The actor, front portrait

Portrait of a twenty year old man, short dark tousled hair, tired brown eyes, light stubble, olive skin, wearing a faded gray hoodie under a navy work jacket, standing in cold early morning light on an empty street, neutral weary expression, photorealistic, 35mm film quality, natural film grain

R2 · Profile (attach R1 as reference)

Three quarter profile of the same young man from the reference image, same short dark tousled hair, light stubble, same faded gray hoodie under a navy work jacket, looking off frame with a wary expression, cold blue morning light on his face, empty street behind him, photorealistic, 35mm film quality, natural film grain

R3 · Full body (attach R1 as reference)

Full body shot of the same young man from the reference image, short dark tousled hair, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack over one shoulder, standing alone in the middle of an empty city street at dawn, cold haze, photorealistic, 35mm film quality, natural film grain

R4 · Style frame — the color grade anchor

Cinematic still frame with no people: an empty city avenue at 6 AM, cold steel blue haze, desaturated concrete grays, one distant warm amber traffic light, low morning sun barely rising between buildings, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights

R5 · Bedroom

A small modest bedroom at dawn, single bed with rumpled blankets, an analog alarm clock on a shelf above the headboard showing 6:00, cold blue light through thin curtains, one warm tungsten bedside lamp, clothes over a chair, lived in and slightly untidy, photorealistic, 35mm film quality, natural film grain

R6 · The empty main road

Wide shot of a deserted city main road at early morning, no people, no moving cars, a lonely bus stop with an empty bench, traffic lights changing for nobody, scattered parked cars, cold haze, low sun, photorealistic, 35mm film quality, natural film grain

R7 · Deserted city center

Extreme wide shot of a large deserted city center at morning, empty boulevards, silent storefronts, a large closed chain store with glass doors, not a single person anywhere, still air, cold blue and gray palette with one amber shop sign, photorealistic, 35mm film quality, natural film grain

R8 · Police station with patrol car

Street view of a small city police station at morning, glass entrance doors ajar, interior lights still on, a white and blue police patrol car parked outside with windows rolled down, empty street, cold morning light, photorealistic, 35mm film quality, natural film grain

R9 · The café

Interior of a large famous city cafe, warm amber pendant lights on, gleaming espresso machine behind a long wooden counter, empty chairs and tables perfectly arranged, morning light through tall windows, no people, inviting and eerie at once, photorealistic, 35mm film quality, natural film grain

R10 · The dealership

A luxury car dealership glass showroom on a city street at morning, expensive sports cars behind tall glass, one sleek silver convertible with an open roof parked near the exit, empty street reflected in the glass, cold light with warm showroom interior, photorealistic, 35mm film quality, natural film grain
Note on aspect ratios: generate the actor images in portrait (3:4) — more facial pixels means a stronger identity lock — and the locations in 16:9. Reference images never dictate the video's aspect ratio; that's set at generation time.

Step 4 · The prompt anatomy

Every video shot uses the same skeleton. First, references are bound by slot number — and this matters: load exactly five images per shot, in the same order every time (the actor's three images, the style frame, then that scene's location). Our first attempt loaded all ten references at once and the generation failed outright — Seedance accepts at most nine, and fewer, relevant ones work better anyway. Then comes the identity block, then the timestamped beats, then the sound list.

The shared identity block (opens every shot prompt)

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.
Settings for every shot: Seedance 2.0 · 16:9 · 480p · Generate Audio ON · duration per shot. About 2 credits per second at 480p.  

Step 5 · The fourteen shots — every prompt

Paste the identity block first, then the shot's timeline below it. The location to load as [Image 5] is noted per shot.

Shot 01 · The alarm · 10s · location: bedroom (no lip needed — he sleeps)

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the small bedroom at dawn with the analog alarm clock on the shelf above the bed; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] Static tripod frame, close on the analog alarm clock above the bed reading 5:59; the room is cold blue, the bedside lamp off; the second hand ticks, the only sound.
[3s] The clock hits 6:00 and rings violently, metal bells hammering; below the frame edge a young man stirs under blankets.
[6s] EMRE's hand rises into frame and silences the clock; the ring cuts to dense silence; he sits up slowly on the bed edge, face in shadow, shoulders heavy with sleep.
Sounds: clock tick, harsh alarm bells, fabric rustle, then near total silence with faint room tone.
Shot 01 as rendered: the hand, the clock, the cold blue quiet

Shot 02 · Leaving · 12s · location: bedroom

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's starting location — the small bedroom at dawn; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and locations throughout.

[0s] Static frame by the bedroom door: EMRE pulls the navy work jacket on over his gray hoodie, grabs his small backpack from the chair, checks his phone once and pockets it.
[5s] He exits through the door; the frame holds on the empty room for one beat, the thin curtains breathing once in a draft.
[7s] Exterior, narrow residential street at dawn: one slow smooth tracking shot behind him as he walks between parked cars toward the main road, his breath faintly visible in the cold, the street completely still.
Sounds: zipper, keys jingling once, a door closing and echoing slightly too long, his footsteps on pavement — and nothing else, no birds, no distant traffic.

Shot 03 · The reveal · 15s · location: main road

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the deserted city main road at early morning with the lonely bus stop and traffic lights; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE steps from a narrow side street onto the main road mid stride and slows to a stop; medium shot from the front, his eyes scanning left to right across the emptiness.
[4s] Slow push in on his face: brow tightening, a small swallow, breath held — the body language of a man realizing something is deeply wrong.
[8s] Cut to what he sees: one very slow pan across the road from [Image 5] — no cars moving, no pedestrians, the empty bus stop bench, a traffic light cycling green to amber to red for nobody.
[13s] Back on him from behind, small in the frame; he turns fully around once, alone in the middle of the road.
Sounds: wind moving over open asphalt, the faint electric buzz and relay click of the traffic light changing, one step scraping grit, his slow exhale.

Shot 04 · The bus that never comes · 12s · location: main road

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the deserted main road with the lonely bus stop bench; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] Static wide: EMRE sits alone on the bus stop bench, elbows on knees, the empty road stretching behind him in both directions.
[5s] Closer static frame: he leans out to look down the road for a bus that is not coming; nothing moves; he checks the time on his phone, thumb tapping the screen twice.
[9s] He exhales through his nose, stands, shoulders the backpack and walks out of frame; the bench sits empty; hold two beats on the vacant stop.
Sounds: bench creak, jacket rustle, wind, a loose metal sign panel ticking against its post — no traffic, no voices. 
Shot 04: waiting for a bus that will never come

Shot 05 · The silence grows · 15s · location: main road

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the deserted city streets at early morning; match the look exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] Slow smooth tracking shot alongside EMRE walking past shuttered shops, his reflection sliding across the dark glass.
[5s] Static insert: a pedestrian crossing signal blinking its green walking man to a completely empty crosswalk.
[8s] High wide angle from above the intersection: one small figure crossing a vast empty junction diagonally, the scale of the city pressing down on him.
[12s] Ground level behind him: he stops, turns in place scanning the windows above; nothing moves anywhere.
Sounds: his lone footsteps multiplying in echo, the crossing signal relay clicking, a deep hush like the city holding its breath.
Shot 05: the high wide — one man against a city built for a hundred thousand

Shot 06 · The bicycle · 15s · location: main road

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the wide deserted main road; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE notices an abandoned bicycle lying on its side by the curb, its front wheel slowly spinning to a stop; he approaches, rights it, tests the brake lever twice.
[6s] He swings a leg over and pushes off; one smooth tracking shot alongside him gaining speed down the empty road, jacket fluttering, the first hint of ease entering his body.
[11s] Wide from the center of the road: the tiny cyclist riding the middle line of a six lane avenue that belongs entirely to him.
Sounds: freewheel clicks, chain hum, tires on asphalt, wind rising with his speed.
Shot 06: the avenue belongs to him

Shot 07 · The dead city center · 12s · location: city center

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the large deserted city center with silent storefronts; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE coasts into the city center on the bicycle and brakes to a slow stop; he dismounts, letting the bike lean against his hip.
[5s] Slow orbit around him as he turns, head tilting back at the silent towers and dark storefronts; his hands hang loose, mouth slightly open — disbelief carried in the body.
[9s] Extreme wide: one small figure and one bicycle in an empty boulevard built for a hundred thousand people.
Sounds: a single sheet of paper skittering across pavement, wind funneling between buildings, his breathing.
Shot 07: downtown, population one

Shot 08 · Locked doors, dead phones · 15s · location: city center

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the city center with the large closed chain store with glass doors; match it exactly. He never speaks. All emotion is physical: posture, hands, breathing, eyes. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE jogs the last steps to the glass doors of the large chain store and pulls the handles; they hold firm, rattling; he cups his hands against the glass and peers in at dark silent aisles.
[6s] Medium shot: he pulls out his phone, dials, holds it to his ear; his eyes move while it rings and rings unanswered; he lowers it, dials another number, waits again.
[12s] Close: the phone descends slowly from his ear; his thumb hovers over the screen, then the hand drops to his side, screen still glowing against his jeans.
Sounds: glass doors rattling, phone dial tones, ringback tone repeating unanswered, the tiny click as he ends the call, wind. 

Shot 09 · The police station · 15s · location: police station

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers, small backpack. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the small city police station with glass doors and lights on inside; match it exactly. He never speaks except one shouted hello. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE pushes through the unlocked station doors; slow steadicam follow behind his shoulder down a corridor where fluorescent lights hum over empty desks, jackets still hanging on chair backs, a coffee mug abandoned mid desk.
[6s] He stops in the middle of the empty bullpen and shouts a hello; the shout comes back to him as an echo off the tile; he waits, chest rising and falling.
[11s] Slow pan across the silent room, ending on him small in the doorway, one hand resting on a counter, knuckles whitening.
Sounds: fluorescent hum, his single shouted hello and its echo, a wall clock ticking, chair wheels shifting a centimeter as he brushes past a desk.

Shot 10 · The siren · 15s · location: police station

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the police station exterior with the white and blue patrol car parked outside, windows rolled down; match it exactly. He never speaks. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] Outside the station: EMRE circles the parked police car, sees the open window, hesitates with one glance over each shoulder, then opens the door and slides into the driver seat; the key hangs in the ignition.
[5s] Interior close: his hand turns the key; the engine catches — and the siren erupts at full volume; he flinches hard, hands flying across unfamiliar switches and buttons on the dash.
[9s] One static extreme wide of the deserted street: the siren wail rolling unanswered between the buildings, the loneliest sound in the world.
[13s] Interior: he finds the switch; the siren dies mid howl; he sits back, exhales, and a first involuntary half smile crosses his face.
Sounds: door hinge, engine start, full police siren blast, the echo decaying over the empty city, sudden silence, one small laughing exhale.
Shot 10: the loneliest sound in the world

Shot 11 · The city is his · 12s · location: city center

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the wide boulevards of the deserted city center; match it exactly. He never speaks. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] Smooth tracking shot alongside the white and blue police car rolling unhurried down the empty boulevard, EMRE's elbow resting out the open window — ownership entering his posture for the first time.
[5s] Interior over his shoulder: fingers drumming the wheel; in the rear view mirror his eyes have lost their fear; he takes a corner wide and lazy because no one can stop him.
[9s] Wide from a rooftop: the lone patrol car drifting through the dead city like a toy on an empty table.
Sounds: engine, tires on asphalt, wind through the open window, one contented exhale.

Shot 12 · The café dream · 15s · location: café interior

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, warm amber interior accents, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the grand famous city cafe interior with the long wooden counter and espresso machine; match it exactly. He never speaks aloud but mouths playful words. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] EMRE walks behind the long counter of the empty grand cafe, runs his palm along the wood, then works the espresso machine: portafilter knock, grinding, tamping — his hands practiced and happy.
[6s] The steam wand hisses; he swirls milk into a latte with complete focus, then looks up at the empty room and — playing — takes an invisible customer's order with a small bow and a genuine grin, sliding a second cup down the counter to no one.
[12s] He raises his own cup in a salute to the empty chairs.
Sounds: grinder, portafilter knock, steam hiss, cup on saucer, a cup sliding along wood, his quiet chuckle echoing in the empty room. 
Shot 12: living the dream, audience of zero

Shot 13 · The idea · 15s · SPECIAL: six references — café as [Image 5], dealership as [Image 6]

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is the cafe he sits outside of; [Image 6] is the luxury car dealership glass showroom across the street with the silver convertible — match both exactly. He never speaks. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and locations throughout.

[0s] Exterior cafe: EMRE sits at a sidewalk table with his latte, feet up on a second chair, surveying the deserted city like a king surveying a kingdom; slow push in as calm settles over his face.
[6s] His eyes catch something across the street; rack focus from his face to the reflection in the cafe window: the dealership glass showroom from [Image 6], the silver convertible gleaming inside.
[11s] Back on his face: a smile arriving slowly, one eyebrow lifting; he sets the cup down with a decisive click and stands.
Sounds: cup on metal table, a distant flag rope clinking against a pole, his chair scraping back on the pavement.

Shot 14 · The fall — the ending · 15s · location: dealership

[Image 4] [Image 3] [Image 2] are EMRE — maintain his exact identity in every shot: twenty year old man, short dark tousled hair, tired brown eyes, light stubble, faded gray hoodie under a navy work jacket, black jeans, worn white sneakers. [Image 1] is the color grade — steel blue shadows, concrete gray, one warm amber accent, desaturated, ARRI ALEXA aesthetic, 35mm film quality, natural film grain, halation on highlights. [Image 5] is this scene's location — the luxury dealership exit onto the empty street, with the silver convertible with open roof; match it exactly. He never speaks. All emotion is physical. Audio is diegetic only — the sounds listed below, no music. Maintain consistent facial features, wardrobe and location throughout.

[0s] The silver convertible rolls slowly out of the dealership lot with its roof open, EMRE at the wheel wearing the grin of a man inside a dream; he brakes near the exit and leans down to the stereo, pressing buttons that do nothing.
[6s] A distant scream cuts through the silence — high, human, falling in pitch; his head snaps up; through the open roof, high above, a small dark shape is dropping fast against the pale sky.
[11s] Hard cut to a static wide shot of the street from a distance at the exact moment of one massive metallic impact sound; the frame holds one second on the still street; cut to black.
[13s] Black screen, silence — then only a distant car alarm beginning to wail. Hold the black to the end.
Sounds: engine idle, stereo button clicks, the distant scream growing closer, a rush of wind, one massive metal impact exactly on the cut, dead silence, a far away car alarm.
Shot 14: the last seconds before the cut to black
Why the impact is off screen: a scream, a falling shape, a hard cut on the impact sound, black. It is stronger filmmaking (the audience's imagination does the work) and it keeps the prompt clear of content filters. Never prompt the body or the crash itself.  

Step 6 · The edit

Download all fourteen clips named shot01 to shot14 and cut them back to back with hard cuts — not crossfades. Cuts are film grammar; dissolves read as montage. We book ended the film with a fade in title card (06:00) and an end card (EPISODE ONE) over black, both generated with a single ffmpeg command each, then concatenated everything:

ffmpeg -i title.mp4 -i shot1.mp4 ... -i shot14.mp4 -i endcard.mp4 \
 -filter_complex "[0:v][0:a][1:v][1:a]...concat=n=16:v=1:a=1[v][a]" \
 -map "[v]" -map "[a]" -c:v libx264 -crf 18 -c:a aac episode-one.mp4

Any video editor does the same in a drag and drop timeline. Because every clip carries its own diegetic sound, the film works with no score at all — though a single low ambient drone under the whole timeline, warming as the film warms, is a worthy 5 credit addition from Voice Studio.

Budget, lessons, and what broke

Item Credits
Reference pack — 10 images, High quality 40
14 shots at 480p (10 to 15 seconds each) about 380
Complete 3 minute 21 second episode about 420 credits

What broke and what we learned. Our first generation failed instantly — we had loaded all ten reference images, and the model accepts nine at most. The fix became the best practice: five references per shot, always in the same order, with only the location swapping. Identity held across all fourteen shots because the actor's three reference angles rode along on every generation. And the single most repeated phrase in the whole production — "maintain consistent facial features, wardrobe and location throughout" — earned its place in every prompt.

The whole stack is one product: Quick Image Generator casts your actor and builds your sets, Pro Studio with Seedance 2.0 shoots coverage with sound, and the Template Library teaches the prompt craft. The only ingredients you bring are a story and an evening.

Episode Two is already falling toward us. Open Zoviz Studio and start yours.

Recibe boletines exclusivos

Suscríbete ahora para recibir las últimas novedades directamente en tu correo. ¡No te lo pierdas!