Google Omni Prompt Generator

Generate a unified text + video + audio output: A 20-second documentary sequence following a traveler arriving in Lisbon for the first time. Shot 1 (5s): Exterior of Santa Apolónia station, morning light, pigeons, tram in background. Shot 2 (7s): The traveler exits the station and stops — the city opens up before them, the Tagus visible in the distance. Shot 3 (8s): Close on their face, a quiet smile. Narration: 'Some cities tell you something the moment you arrive. Lisbon told me I'd been here before.' Lip sync on the narration in shot 3. Ambient audio throughout: station echoes, tram bell, river breeze, morning city hum. Cinematic documentary quality.

A complete 30-second brand narrative in one generation. Opening (0–8s): aerial view of a craftsperson's workshop at dawn, tools laid out precisely. Middle (8–20s): close-up sequence — hands at work, material being shaped, focus and care visible in each motion. Closing (20–30s): the finished object — a leather bag — placed on a table, the maker steps back. Voice-over on closing: 'Made once. Made right.' Soundtrack: spare acoustic guitar, no percussion. Audio locks to every visual beat. Premium brand aesthetic.

A multi-modal educational sequence on ocean acidification. Generate: (1) A text explanation of the mechanism in 3 sentences; (2) A 15-second visual sequence: time-lapse of coral bleaching over 20 years compressed to 8 seconds, then a split-screen showing healthy vs. acidified ocean chemistry; (3) A narration track: 'Every year, oceans absorb a quarter of the CO₂ we emit. The chemistry changes. The life changes with it.' Narration is lip-synced to a presenter in frame in the final 5 seconds. Educational documentary quality.

A 45-second music video sequence for an indie folk track. Location 1 (0–15s): a musician plays acoustic guitar alone in a wheat field, late afternoon, no other sound. Location 2 (15–30s): the same musician in a rain-soaked city underpass at night, neon reflections, same guitar part. Location 3 (30–45s): rooftop at sunrise, the full band joins for the chorus. Audio: the instrumental track runs continuously across all three locations; the acoustic transitions to full band precisely at the location-3 cut. The artist's lip movements on the vocal lines are synchronized throughout.

A 20-second broadcast news segment. Generate: anchor in studio says: 'Scientists have confirmed the discovery of a new deep-sea ecosystem near the Azores — the largest found this decade.' Lip sync precise. Cut to: underwater drone footage of bioluminescent creatures at 3,000 metres — 8 seconds of new footage. Cut back to anchor: 'The discovery was made last week and is being described as significant for marine biology.' Ambient: studio neutral, underwater sequence is total silence except for low pressure ambience. Broadcast news quality.

The opening 30 seconds of a short film. Scene establishes: a woman in her 40s sits at a kitchen table at 6 AM. A letter is in front of her, unopened. She hasn't slept. She finally opens it, reads, and says quietly to herself: 'Well. That's it then.' Lip sync on the line. Camera: starts wide, pushes slowly to medium as she reads, holds on her face for the line. Audio: refrigerator hum, dawn birdsong through a window slightly ajar, the sound of the envelope opening, her words are the only human sound. 30 seconds.

A complete tech product explainer in one pass. Generate: a presenter in their 30s, smart casual, stands in front of a large display screen. They say: 'This changes how you store energy at home — permanently.' The screen behind them shows: first a diagram of the battery system (8s), then a real-world installation clip (8s), then usage statistics (8s). Lip sync precise throughout. Audio: presenter voice is dominant; screen content has subtle ambient audio — the installation clip has appropriate outdoor sounds. Clean, modern tech presentation.

A 25-second food documentary sequence. A chef in their 60s, Oaxaca kitchen, demonstrates making mole negro. They narrate: 'The chilhuacle negro — you must toast it until the smoke is just this side of bitter. No recipe tells you when. You learn it from your hands.' Lip sync on narration. Close-up shots of the chili toasting, the smoke rising, the colour change. Audio: the cooking sounds are layered carefully — the hiss and crackle of the chili, the kitchen background, the chef's voice over all of it. Documentary warmth. 25 seconds.

Generate a unified 30-second B-roll package for a Japan travel video. Include: Shibuya crossing at rush hour (8s), bullet train passing Mount Fuji at dawn (8s), hands wrapping onigiri in a convenience store kitchen (7s), cherry blossoms falling in slow motion in Maruyama Park (7s). Audio: each location has appropriate ambient — crossing cacophony, train wind-rush, the snap of plastic wrap, wind through petals. No voiceover. No music. Pure location audio for editorial flexibility. 4K quality.

A 20-second podcast segment with visual generation. Two hosts, mid-shot, studio setting with plants and acoustic panels. Host 1 says: 'The thing nobody tells you about learning a second language is that it changes how you think in your first.' Host 2 responds: 'Completely — there's a Portuguese word, saudade, that English speakers feel after they learn it.' Both lip-syncs are precise. Audio: studio-quality podcast mic sound, slight room tone, no music. Natural conversation pacing. 20 seconds.

A 20-second architectural walkthrough of a completed building. Generate: exterior approach along a stone path at golden hour (5s), entry through a pivot door into a double-height atrium (5s), the atrium in full — natural light through a north-facing clerestory, a single tree growing through the floor (10s). Audio: footsteps on stone, the solid sound of the pivot door mechanism, then the acoustic shift to the atrium interior — a gentle spaciousness, nothing harsh. Architectural photography quality. No voiceover.

A 25-second historical drama fragment set in 1940s London. An officer in uniform meets a woman on a train platform. He says: 'This is the last train for six months. I don't know when I'll be back.' She replies: 'Then I'll be here in six months.' Both lip-syncs are precise. Camera: medium two-shot for dialogue, brief close on her face at her line. Audio: the station — steam, crowd noise, a loudspeaker announcement, the train whistle at 22 seconds. Period-accurate sound design. BBC historical drama quality.

A 20-second motivational fitness reel. Visual sequence: pre-dawn alarm (2s), running shoes laced in the dark (3s), a runner exiting an apartment building into empty dawn streets (5s), a 10km run condensed to 5 seconds of landmarks and effort, finishing back at the same door (5s). Voice-over: 'The city is yours before anyone else wakes up.' Audio: alarm buzz, then silence, then footsteps building into the run — city ambience growing as the run progresses, voice-over over the finish. Energetic, not aggressive.

The 15-second intro sequence for an educational series on climate science. Generate: animated globe with CO₂ concentration heat-map from 1850 to 2026 in 6 seconds; cut to photorealistic timelapse of a glacier retreating over 40 years in 4 seconds; cut to the series host, mid-shot, who says: 'This is what we know. This is what we can still change.' Lip sync precise. Theme music: spare, modern, not alarming — 4 chords held under the sequence. Clean educational broadcast quality.

A 30-second wedding ceremony highlight. Shot sequence: wide of the venue — a converted barn, late afternoon light through high windows (5s). Close on the couple's hands being joined by the officiant (5s). The groom turns to the bride and says: 'I have been looking for you my whole life.' Lip sync precise. The bride's reaction — a laugh and tears simultaneously — captured in medium close-up for 8 seconds. Final 12 seconds: wide shot of the room, guests standing, natural applause. Audio: room acoustic, the spoken line dominant, the applause warm and full.

A 25-second wildlife sequence on wildebeest migration. Aerial wide showing the herd scale — 400,000 animals, dust rising (8s). Ground-level tracking shot alongside the herd at speed (8s). A crocodile strike at a river crossing — one wildebeest in, one escaping (9s). Audio: the entire sequence is sound-designed precisely — the herd rumble from the air, the ground-level thunder of hooves, the river crossing explosive splash and crocodile thrash. No narrator. Pure nature documentary audio. BBC Earth quality.

A 20-second public service announcement on urban loneliness. Visual: a montage of single people in public — a man eating alone in a busy restaurant, a woman on a packed Tube train looking at nothing, an elderly man waving at a closed window. Each portrait is 5 seconds. Voice-over at the end: '37% of adults say they feel lonely most of the time. Most of them are surrounded by people.' Audio: the environments are full of noise — restaurant buzz, train sound, street ambience — which makes the isolation more acute. Public awareness documentary quality.

A 25-second founder story sequence. The founder, 30s, speaks directly to camera: 'My daughter was diagnosed at eight months. There was no early detection tool that worked. So we built one.' Cut to: a lab sequence — researchers at work, screens with data, a small device being tested on a bench (10s). Cut back to founder: 'It caught 94% of cases in trial. We're scaling now.' Both dialogue sections are lip-synced precisely. Audio: founder voice clean and natural; lab sequence has appropriate ambient lab sound. Confidence, clarity, no music.

A 20-second cinematic landscape sequence in Iceland. Shot 1: Aerial over the Vatnajökull glacier at blue hour — infinite white, steam vents, scale impossible to grasp (8s). Shot 2: ground-level of lava field from the 2021 Fagradalsfjall eruption — frozen black waves, tiny geologist figure for scale (7s). Shot 3: the Northern Lights beginning, 9 PM in September — the first green arc appearing over a farmhouse (5s). Audio: the glacier — wind, total cold silence with faint creak; lava field — boot on volcanic rock, wind; Lights — silence and the faint sound of a window opening. No music. Pure atmospheric.

A 30-second multimodal AI capability demonstration video. A presenter stands before a display. They say: 'You give it a photograph, a spoken description, and a video reference. It understands all three simultaneously.' The screen shows: a user uploading an image (5s), speaking a description (5s), dropping a video reference clip (5s). Then the output appears on screen — a generated video sequence matching all three inputs precisely. Presenter: 'One prompt. Three inputs. One output.' Lip sync on both dialogue lines. Clean tech demo aesthetic, modern sans-serif UI visible on screen.