Capabilities

You, talking to the camera — without a camera, and without the look that gives AI videos away.

A faceless video for the price of an edit: the voice goes on — the speaker's own, or a script read by ElevenLabs — while stock footage and photos fill the screen, a new shot every…

Your story or a customer's: how it began, what changed, what you learned.

A product or service on every slide, and how to order at the end.

A short list of tips, one per slide. The post people save.

What everyone believes and what is true: a myth, then its truth.

The result of a job: before, what you did, after. Best with your own pictures.

A process in order, one step per slide. For how-tos and questions that keep coming.

A finished Instagram carousel from a short brief: designed slides, a caption and hashtags.

15 s: a full-screen hook, a split screen with the work and a counting price, numbered steps on the footage, a call to action

15 s: a full-screen opening with a round gold badge, a split screen with the product and the price, a finale with a second badge

15 s: a full-screen opening, slanted bars with the place and three icons, then a huge 'free' or price with a gift

15 s, slow and soft: an elegant opening, a rounded cream panel with a list and the price, a finale under a cream arch

A person from the business beside a big curved white panel: a local hook, a handwritten line, a huge price, a gift ribbon and three benefits on a brush stroke

A tall photo beside a black-and-gold panel: logo, a framed second photo, a bold headline, three gold icons and the price

A white arch with the headline over a bright photo, a blue price box with a gift, a round stamp and two benefit pills

A cut-out person over slanted yellow and white bars on navy: the activity, the place, a huge 'free' or price and a button

A finished marketing banner for a business from a short brief: a square picture or a 15-second square video, with the business's logo and photos or all made with AI.

The trending CapCut Reel, modelled frame by frame on two reference videos: no subtitle line — every word lands on the frame it is said, in small cards that build up and clear…

The documentary short: the speaker tells a story and the engine draws its facts — a year that counts up, view counts that stay tiny, a wall of videos filling, a count against the…

The Hebrew product-review Reel, measured frame by frame on a creator's chair review: the price counts up in huge green italic with the hook line typed in under it, a whip into the…

A YouTube video becomes a vertical reel, on your rights: search YouTube (Creative Commons only, when the video is not yours) or bring a link, and Gemini watches it where it is and…

The top of the range: the takes are cut and joined into one story, then Gemini watches and listens to the result — faces, pauses, what the words cannot say — and directs the whole…

The AI director at a third of the price: the same recipe — takes joined, Gemini watches and directs, a score timed to the story, sound design, kinetic words — with three short Veo…

The simple finished video: Gemini watches the clip and writes a title that stops the scroll and a brief for the music, ElevenLabs makes a track that sits under the voice, and one…

The talking-head format of a CapCut creator with 300K followers, measured frame by frame on his Reels: a designed title over the first sentence (a heavy line, a serif line and a…

A CapCut creator's label captions, measured on his Reels: every 2–3-word phrase lands on a solid coloured label at the chest (his orange serif labels, the blue italic ones, purple…

The format of a CapCut creator's most viral Reel (1.1 million views): a board of pill rows that fills with the items the moment the speaker names them — a checklist when he lists…

A CapCut creator's listicle, measured on his Reels: every point the speaker makes opens with a white flash, its number on a yellow starburst and its name in big yellow serif at…

The kinetic CapCut edit of a talking head with real B-roll instead of generated clips: Gemini Flash-Lite reads the words and names the moments a picture helps, the stock is…

The everyday short: a punchy hook, big green-highlight captions, a hard silence cut, punch-in zooms on the question, the number and the punchline with a whoosh on each, and an…

A list or a process: a box hook, news-bar captions, one free icon card per step and a calm music bed.

Generated B-roll: three 4-second Veo clips and two Nano Banana Pro stills cut in over the talking head on the moments that need a picture, a focus hook, elegant serif captions…

A selling clip: a counting-number hook, pop captions, Nano Banana Pro product shots, icons on the proof points, a kinetic end card with the price (or the product, when no price is…

Captions synced to the speech in the Classic look — Rubik 800, highlight animation, 5 words at a time.

Captions synced to the speech in the Hormozi look — Heebo 900, highlight animation, 4 words at a time.

Captions synced to the speech in the Karaoke look — Secular One 400, highlight animation, 6 words at a time.

Captions synced to the speech in the One word look — Suez One 400, word animation, one word at a time.

Captions synced to the speech in the Marker look — Heebo 900, highlight animation, 5 words at a time.

Captions synced to the speech in the News look — Frank Ruhl Libre 700, highlight animation, 8 words at a time.

Captions synced to the speech in the Neon look — Varela Round 400, highlight animation, 6 words at a time.

Captions synced to the speech in the Quiet look — Assistant 600, highlight animation, 8 words at a time.

Captions synced to the speech in the Pop look — Karantina 700, pop animation, 3 words at a time.

Captions synced to the speech in the Typewriter look — IBM Plex Sans Hebrew 500, typewriter animation, 8 words at a time.

Captions synced to the speech in the Bounce look — Rubik 900, bounce animation, 4 words at a time.

Captions synced to the speech in the Underline look — Heebo 900, highlight animation, 6 words at a time.

Captions synced to the speech in the Elegant look — Noto Serif Hebrew 600, highlight animation, 7 words at a time.

Captions synced to the speech in the Handwritten look — Amatic SC 700, highlight animation, 4 words at a time.

Captions synced to the speech in the Comic look — Varela Round 400, pop animation, 4 words at a time.

Captions synced to the speech in the Sticker look — Heebo 900, highlight animation, 5 words at a time.

Captions synced to the speech in the Drama look — Secular One 400, highlight animation, 5 words at a time.

Captions synced to the speech in the Cinema look — Heebo 300, highlight animation, 7 words at a time.

Captions synced to the speech in the Long shadow look — Rubik 900, highlight animation, 5 words at a time.

Captions synced to the speech in the Lime glow look — Heebo 900, highlight animation, 5 words at a time.

Captions synced to the speech in the Retro look — Karantina 700, highlight animation, 3 words at a time.

Captions synced to the speech in the Scale look — Heebo 900, highlight animation, 5 words at a time.

Captions synced to the speech in the Bubbles look — Assistant 800, highlight animation, 5 words at a time.

Captions synced to the speech in the Sport look — Suez One 400, highlight animation, 4 words at a time.

An animated title in the Colour box look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Typewriter look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Punch zoom look for the first three seconds — the title moves with force; the same look works on any line mid-video.

An animated title in the Glitch look for the first three seconds — the title moves with force; the same look works on any line mid-video.

An animated title in the Marker look for the first three seconds — one word is marked out; the same look works on any line mid-video.

An animated title in the Word by word look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Box look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Counter look for the first three seconds — one word is marked out; the same look works on any line mid-video.

An animated title in the Shake look for the first three seconds — the title moves with force; the same look works on any line mid-video.

An animated title in the Rise look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Strike look for the first three seconds — one word is marked out; the same look works on any line mid-video.

An animated title in the Focus look for the first three seconds — the words appear on the screen; the same look works on any line mid-video.

An animated title in the Flash look for the first three seconds — the title moves with force; the same look works on any line mid-video.

Types the line on word by word, like a typewriter.

A highlighter bar wipes in behind the line, from the reading side.

The text arrives with a quarter-second of RGB-split jitter, then settles clean.

The text slams in from 160% and lands with a small overshoot.

The line shudders for a moment — a warning, a shock — then settles.

Each line rises from behind an invisible edge, staggered — the clean editorial title.

A word gets struck through and the right word lands in its place in the accent colour.

Letters come into focus while the spacing tightens — the cinematic title reveal.

Each spoken word pops onto the screen on its own, exactly when it is said.

The word being spoken sits on a coloured rounded box that jumps from word to word.

1–3 spoken words per card in rotating display fonts and colours, each card replacing the last on the word.

A white sticker with the key word in red slams in, tilted — the call-to-action label.

A big number counts up from zero and lands on its value, over the video.

A small label on a coloured pill at the top — who is speaking, a place, a chapter.

The line is typed letter by letter in reading order, a thin cursor leading and blinking.

One big word slams in with a solid 3D extrusion — the “BOOM!” of an impact.

Each word flips into place round its axis, one after another — like a split-flap board.

Random characters shuffle and resolve into the real text letter by letter — the hacker reveal.

Letters drop, rise, slide or grow in one by one in reading order — and can keep bobbing in a wave.

One word of a sentence keeps swapping — “we build apps → sites → bots” — sliding through its slot.

The video shows only through huge letters, then the letters zoom through and reveal the whole frame.

Captions where each word fills with colour as it is said — sweeping right to left in Hebrew.

The two shots dissolve into each other, picture and sound.

The picture dips to black (or white) between the shots.

The next shot pushes the last one off-screen — or slides in over it (the push-up cover).

A line or shape wipes the next shot in: straight, diagonal, iris, clock or slices.

Rushes into the outgoing shot and lands zoomed-out on the next one, with motion blur.

A fast blurred swipe carries the viewer to the next shot.

The frame spins round with blur into the next shot.

A few frames of digital breakup across the cut: bands jump, colours split.

A 3–4 frame white flash on every jump cut — makes each cut feel intentional.

A quick dissolve through a burst of warm light that hides the cut.

The next shot bleeds in like ink from a point, or is painted on in brush strokes.

The frame freezes and breaks into shards that fly apart and fall — the video goes on behind them.

Every jump cut becomes a short morph: the speaker seems to shift instead of jumping.

An instant zoom-in on a word or beat — the talking-head emphasis cut.

A slow, smooth push-in (or pull-out) across a sentence — builds tension.

A slow push-in builds to the hit, then the frame snaps back out on it — the recoil of an impact.

An impact shake on a hit or a punchline — or a handheld sway.

A skate-video fisheye bulge on the picture.

Moving hands and heads leave fading copies or light trails behind them.

A slow pan-and-zoom across a still photo — b-roll photos and faceless videos come alive.

A full-page screenshot scrolls by — a smooth eased scroll, or a fast whip with motion blur.

The frame bumps, flashes or shakes on every beat of the music — the beat is found in the track.

Ready grades — warm, cool, vivid, cinematic, teal-orange, black-and-white, vintage, matte — plus fine-tuning.

For a beat the speaker turns black-and-white, dims, blurs or tints — a question, a twist, a flashback.

The red and blue of the picture jump apart for a frame or two — the chromatic shock of a hit.

Organic, moving film grain over the picture.

Two-colour poster look: shadows in one colour, highlights in another.

Everything turns black-and-white except one colour — the blue sky, the brand colour.

Highlights glow — a red film halation round bright edges, or a soft dreamy bloom.

Period looks in one effect: VHS tape, an old CRT, Super 8 home movie or a 2000s digicam with an orange date.

An anamorphic blue streak or a warm glow sweeps across the frame, its ghost rings trailing opposite.

Apply a colour LUT — one of ours (teal & orange, warm film, bleach, night) or your own .cube — at any strength.

A number counts up (or down) to a value — money, followers, years — with an optional progress bar.

A white flash for a frame or two — on a hit, a cut or a punchline; the whole frame or one region.

Everything dims except one area — a face, a line of text, a product — which stays bright.

The hit in one call: the frame pushes in and snaps back, a white flash, a punch word and a boom.

The speaker blurs and dims, and a “comment KEYWORD” sticker slams in on a flash with a deep boom.

Rings of light burst from a point with sparkles — the hit behind a number, a logo or a word.

Comic-book lines burst from a point to the edges for a few frames — the accent under a hit.

Soft light rays turn slowly behind a point — the glory behind a logo, a number or a word.

A thin bar along the top or bottom that fills as the video plays — a retention trick.

A name and title bar slides in from the reading side (who is speaking, a place, a price).

A phone-style notification drops in from the top with a sound — “New order: ₪249”.

An arrow draws itself toward a spot with a label — “look here”.

A hand-drawn circle, box or underline draws itself round something on screen.

Items appear one by one and get ticked ✓ or crossed ✗.

A big quote with quote marks and the author, the words rising in line by line on a clean card.

Confetti bursts, snow falls, sparkles or embers drift over the shot — the celebration beat.

A round loupe magnifies part of the frame.

Bars or a waveform move with the voice — the podcast audiogram.

Warm film light leaks drift over the shot.

Small tiles build one per spoken noun, each line icon drawing itself in, a label under it.

Glowing outline icons pop in one after another — or draw themselves line by line.

A spark appears and an arc draws from one icon to the next, which pops as the line arrives.

One line icon flows into the next — chart into timeline into chat — on a tile or on its own.

A dark navy plate with soft light shafts — a panel across the top or the whole frame — to put graphics on.

An icon on its tile cracks in two: a bright zigzag flashes, the halves slide apart and tilt, then fade to a grey ghost.

A browser window: the address is typed, the page loads, scrolls and zooms into the line that matters.

A terminal window types a prompt letter by letter, then a green status line — the AI at work.

The “watch it work” demo in one call: a terminal types the prompt, a browser opens the page, scrolls and zooms into what matters.

A motion-graphics panel in one call: neon chat bubbles pop, a line draws to a play icon, an icon morphs and a line types under it.

The video shrinks into a phone on a dark plate, a follow button pops under it and a cursor clicks it.

Bars grow, a line draws itself or a donut fills round — with labels and values counting up.

A follow button pops in, a cursor glides over and clicks it: it turns to “following ✓” and a bell rings.

A comments sheet slides up, “comment KEYWORD” slams in, the keyword arrives as a comment and a reply follows.

A slider sweeps between the “before” and the “after” (photos or clips) with a handle and labels.

A photo pops in as a rounded card with a thin border and a soft glow, floats a little and fades out.

App icons fly in one after another on short curved paths, pop to size and settle in a row.

Boxes pop in one after another and arrows draw between them: a flow, a cycle or a timeline.

Emojis burst out of a point, float up or rain down — hearts, fire, claps, 100.

A second video or image in a small window — corner, floating, tilted, with a shadow.

A visual on one part of the frame, the speaker on the other.

The speaker in a circle with a coloured ring over a designed plate, optional words on an arc.

Puts the video inside a phone on a clean background, rising into place.

The speaker moves to the bottom and a dark graphics panel takes the top — the stage for a demo or a diagram.

The video and your clips or photos in a grid — 2 to 6 tiles popping in: a duet, a comparison, a reaction.

Copies of the speaker, cut out and standing behind him — up to 5 of you in one frame. Or the classic shot: film 2–5 takes from a tripod, standing somewhere else each time, and they become one.

A sound effect on every jump cut the engine makes, peaking on the join.

Short-form sounds — whoosh, swish, hit, boom, bell, pop, scroll, riser, glitch, slow-mo — at a moment or a word.

A music bed that sits well under the voice and dips further while someone speaks.

Telephone, radio, megaphone, robot, echo, hall, chipmunk, deep or whisper on the voice — whole video or one phrase.

The TikTok “slowed + reverb” treatment of the music — or the sped-up “nightcore” version.

Plays the whole video faster or slower (0.25×–4×), the voice kept natural.

The picture drops to a low frame rate for a moment — the stutter of a velocity edit.

The picture freezes (or jumps back to an earlier frame) with a flash and a slow push while the sound goes on.

A moment plays backwards with a tape look and its sound reversed — or forwards and back like a boomerang.

Fast, then slow-motion on the moment, then fast again — inside the span, which keeps its length.

One or two huge words at a time, the spoken word popping in yellow, hard jump cuts.

The phrase stays readable, the spoken word lights up inside a coloured box.

Two quiet lines in the lower third, gentle cut, nothing blinking.

Captions plus 4–6 real photographs from open sources, cut in where the sentence needs a picture.

A list or process video: an icon card and a kinetic label for every step.

History, art and culture told over public-domain images with a slow push-in on each.

Captions, music and 4–5 images generated to illustrate the exact sentence being said.

Colour scenes, mega numbers, marker highlights and punch-ins — typography instead of footage.

A selling clip: AI product shots, icon badges, price card, music and a closing CTA.

The whole toolbox: Veo clips, AI images, split screens, kinetic text, music and effects.

Veo 3.1 and Nano Banana Pro throughout, graded, for a brand film that must look expensive.

The same video in another language: bring a YouTube link — your own channel, a Creative Commons video, or one you have permission for; the licence you declare is kept with the job…

Your own video in other languages, in your own cloned voice with your lips redrawn — and every language checked before it reaches anyone.

A long talk becomes a one-minute reel in Hebrew.

A famous talk becomes a two-minute reel in Hebrew, told in your own voice — the way the owner's Jack Ma reel was made.

A talk becomes a 40-second Reel in another language that looks dubbed, not narrated: Gemini picks the strongest seconds, watches the speaker and writes the dub the way a native…

Your own voice, in Hebrew that sounds like a person and not like a machine reading a page — what the owner's own reels were made of, for everyone.