These 20 prompts are built for Seedance's standout trick: it generates synced speech, lip movement, and ambient sound in a single pass, with phoneme-level lip-sync in 8+ languages. Every prompt below puts the exact spoken line in quotes, names the emotion and delivery, then adds camera, lighting, and a plain settings note for aspect ratio, resolution, and length.

The rule that makes dialogue land: keep spoken lines to one or two short sentences per 8-second clip, reuse @Image1 to hold the same face across shots, and add no subtitles, no on-screen text so quoted lines don't burn in as captions. New here? Start with the 35 best Seedance prompts roundup or the guide to prompting Seedance for realistic video.

Advertisement

Presenters & talking heads

Single speaker to camera — anchors, demos, explainers, creators. Use @Image1 to lock the presenter's face across a whole series and keep the line short enough to breathe.

1. Studio news anchor intro

A composed female news anchor in a navy blazer sits at a modern glass news desk, looking straight into the lens. Use @Image1 as the anchor's appearance. She delivers the line with calm, professional authority: "Good evening — here are tonight's top three stories." Medium shot, locked-off camera, subtle slow push-in. Bright even studio key light with a soft blue set glow behind her. Clean broadcast look. Audio: clear studio dialogue, faint room tone. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Why it works: A locked-off medium shot and a single short line give the lip-sync an easy target and read instantly as broadcast.

2. Product demo talking head

A friendly male presenter holds a small white smart speaker up beside his face in a bright minimalist studio. Use @Image1 as the presenter's appearance. He says, with upbeat energy, "Tap once to play, twice to skip — that's the whole thing." Medium close-up, gentle handheld sway, focus holding on his face. Soft high-key lighting, seamless pale-grey backdrop. Modern commercial style. Audio: warm conversational dialogue, quiet studio ambience. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Best for: Short product explainers where the presenter's face needs to stay consistent across a full launch series.

3. Explainer presenter at whiteboard

A confident presenter in a grey sweater stands beside a clean white glassboard in a bright office, gesturing toward it. Use @Image1 as the presenter's appearance. She explains, clearly and encouragingly, "There are three steps here, and the first one is the easy part." Medium shot, slow dolly-in from the side. Soft daylight from a window camera-left, even fill. Crisp corporate-explainer look. Audio: clear instructional dialogue, faint office hum. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Why it works: The gesture cue plus a single instructional sentence keeps pacing natural and the sync tight.

4. Vertical creator selfie update

A young creator films a selfie-style update while walking through a sunlit city street, holding the phone at arm's length. Use @Image1 as the creator's appearance. He says, excited and casual, "Okay, quick update — the thing I promised is finally ready." Vertical close-up, natural handheld bounce, face centered. Bright natural daylight, soft shadows, warm skin tones. Fresh social-video look. Audio: lively conversational dialogue, light street ambience. No subtitles, no on-screen text. (9:16, 1080p, 8s)

Best for: TikTok and Reels talking-head hooks; see more short-form ideas in the social media prompt pack.

Two-person scenes

Two speakers, one clip. Name each person, give each a short quoted line in order, and use a longer duration so the back-and-forth gets room to breathe.

5. Cafe table catch-up

Two friends sit across a small marble cafe table by a sunlit window, coffee cups between them. The woman on the left leans in and says, warmly, "You actually did it — I can't believe you moved." The man opposite grins and replies, "Neither can I, honestly." Medium two-shot, slow arc around the table. Soft warm window light, gentle bokeh behind them. Naturalistic film look. Audio: friendly dialogue, quiet cafe ambience, distant espresso machine. No subtitles, no on-screen text. (16:9, 1080p, 10s)

Why it works: Two very short lines fit inside 10 seconds, and the two-shot keeps both faces on-screen for the exchange.

6. Office desk disagreement

Two colleagues face off across a cluttered office desk under cool fluorescent light. The man in a shirt says, tense and clipped, "We can't ship it like this." The woman folds her arms and answers, firm, "Then we push the date." 0-4s: medium two-shot holding both. 4-10s: slow push toward the woman as she replies. Cool overhead office lighting, hard edges. Tense workplace-drama tone. Audio: charged dialogue, low HVAC hum. No subtitles, no on-screen text. (16:9, 1080p, 10s)

Best for: Scripted workplace scenes; the time-coded segments hand each speaker a clean beat.

7. Doorway goodbye

A couple stands in an open apartment doorway at dusk, one about to leave with a bag. She says, quietly, "Call me when you land." He pauses, then softly, "I always do." Medium shot from inside the hallway, static frame with a faint handheld breath. Warm interior lamp light against cool blue evening outside. Tender, understated cinematic look. Audio: soft intimate dialogue, faint street sounds through the door. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Why it works: The warm-vs-cool light split adds emotion while the whispered lines keep the sync easy.

8. Kitchen cooking banter

Two people cook side by side at a bright kitchen island, one chopping, one stirring a pot. The chopping woman laughs and says, "You always add too much garlic." The man stirring shoots back, playfully, "There's no such thing." Medium two-shot, gentle handheld, slight drift left to right. Warm kitchen daylight, steam rising softly. Cozy lifestyle look. Audio: light-hearted dialogue, sizzling pan, knife on board. No subtitles, no on-screen text. (16:9, 1080p, 10s)

Best for: Lifestyle and food content where natural banter plus kitchen SFX sells the scene.

Monologues & emotion

One person, one feeling. Push the emotion in your delivery cue and let a slow move on a close-up carry the weight.

9. Rain window confession

A young woman sits alone by a rain-streaked window at night, forehead nearly touching the glass. Use @Image1 as her appearance. She murmurs, on the edge of tears, "I don't know how to tell them I'm scared." Slow push-in to a soft close-up. Low, moody light from a single lamp, cool blue rain glow on the glass. Melancholic cinematic look, shallow depth of field. Audio: fragile whispered dialogue, steady rain on the window. No subtitles, no on-screen text. (16:9, 2K, 8s)

Why it works: The slow push and single lamp isolate the face so the emotion in the delivery cue reads clearly.

10. Locker room pep talk

A team captain in a sweat-soaked jersey stands in a dim locker room, jaw set, breathing hard. Use @Image1 as the captain's appearance. He says, low then rising with fire, "We've been counted out all season — not tonight." Slow push-in on a tight close-up. Hard side light from an overhead bank, deep shadows. Gritty sports-drama look. Audio: intense determined dialogue, distant crowd rumble through the wall. No subtitles, no on-screen text. (16:9, 2K, 8s)

Best for: High-emotion single-line moments where the delivery arc from low to loud drives the shot.

11. Late-night voicemail

A man sits on the floor of a dark bedroom, phone pressed to his ear, city lights soft behind the blinds. Use @Image1 as his appearance and @Audio1 as the dialogue and voice track. He says, tired and honest, "Hey — I just wanted to hear your voice." Static medium close-up, faint handheld sway. Low warm practical light from a bedside lamp, deep surrounding dark. Intimate late-night look. Audio: quiet weary dialogue, faint distant traffic. No subtitles, no on-screen text. (9:16, 1080p, 8s)

Why it works: @Audio1 pins a specific voice while the quoted line keeps the words and lip-sync locked.

12. Mirror self pep-talk

A nervous woman grips a bathroom sink and stares at her own reflection before an interview. Use @Image1 as her appearance. She whispers, steadying herself, "You've done harder things than this." Close-up shot framed through the mirror, slow subtle push-in. Cool morning light from a small window, soft fill on her face. Grounded, honest look. Audio: hushed self-directed dialogue, faint tap drip, quiet room tone. No subtitles, no on-screen text. (9:16, 1080p, 8s)

Best for: Relatable "self-talk" content; the mirror framing gives a natural single-speaker composition.

Advertisement

Narration & voiceover

Here the speaker is off-screen. Write the voiceover line in quotes, describe the visuals it plays over, and use @Audio1 to keep one narrator's voice consistent across a series.

13. Documentary voiceover landscape

Sweeping aerial over misty green mountain ridges at dawn, clouds pooling in the valleys. A calm off-screen narrator says, measured and warm, "For a thousand years, no road ever reached this valley." Slow drone push forward over the peaks. Soft golden dawn light breaking through mist. Epic nature-documentary look. Audio: unhurried narrator voiceover, gentle wind, faint birdsong. No subtitles, no on-screen text. (16:9, 2K, 10s)

Why it works: The voice is off-screen, so there's no lip-sync to manage — the line just needs to fit the shot's length.

14. Brand voiceover product montage

Clean studio montage of a matte-black water bottle rotating, capping, and beading with condensation. Use @Audio1 as the brand voice track. A confident off-screen narrator says, "Built for the long days — and the longer nights." 0-4s: slow turntable rotation. 4-8s: macro push on water droplets. Bright soft key light, subtle rim glow, seamless white sweep. Premium commercial look. Audio: polished brand voiceover, soft ambient pad. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Best for: Ad voiceover; pair with the roundup for matching product-shot prompts.

15. Audiobook narrator study

A candlelit wooden study with an open book, drifting dust, and a rain-lashed window behind. A warm off-screen narrator reads, slow and rich, "She opened the letter she had waited nine years to send." Slow dolly across the desk toward the book. Flickering candle light, deep amber shadows. Atmospheric literary look. Audio: mellow narrator voiceover, crackling candle, soft rain. No subtitles, no on-screen text. (16:9, 1080p, 10s)

Why it works: Atmosphere carries the shot while a single sentence of narration sets the mood without crowding the clip.

16. Museum audio-guide walkthrough

Smooth tracking move through a quiet museum gallery past framed paintings and marble busts. A gentle off-screen guide says, informative and calm, "To your left, the oldest piece in this collection." Steady slow tracking shot gliding forward. Soft even gallery lighting, warm spotlights on the art. Refined educational look. Audio: soothing audio-guide voiceover, faint footsteps and hushed room tone. No subtitles, no on-screen text. (16:9, 1080p, 10s)

Best for: Explainer and tour content where a moving camera pairs with steady narration.

Accents & other languages

Seedance lip-syncs 8+ languages natively. Write the spoken line in the target language, then name the accent or language in the delivery cue so the mouth shapes and voice match.

17. Spanish street-food vendor

A cheerful street-food vendor flips tacos at a bustling night market stall, steam rising from the griddle. Use @Image1 as the vendor's appearance. In lively Mexican-accented Spanish he calls out, "¡Vengan, están recién hechos!" Medium shot, gentle handheld, warm focus on his face. Golden stall bulbs and neon signage, rich warm color. Vibrant documentary look. Audio: energetic Spanish dialogue, sizzling griddle, market crowd. No subtitles, no on-screen text. (9:16, 1080p, 8s)

Why it works: Naming the accent guides both the voice and the phoneme-level lip-sync for the Spanish line.

18. Japanese train announcement

A polite station attendant in uniform stands on a clean Tokyo train platform as a train slows behind her. Use @Image1 as her appearance. In crisp, courteous Japanese she announces, "まもなく電車が到着します。" Medium shot, static frame with a slight lean-in. Bright fluorescent platform light, cool clean tones. Orderly modern look. Audio: clear polite Japanese dialogue, approaching train hum, soft chime. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Best for: Localized announcements and signage-free scenes in non-Latin scripts.

19. French bakery welcome

A warm baker in a flour-dusted apron greets a customer from behind a Parisian bakery counter stacked with croissants. Use @Image1 as the baker's appearance. In friendly French she says, "Bonjour, qu'est-ce qui vous ferait plaisir ?" Medium shot, slow gentle push-in over the counter. Soft warm morning light, golden pastry tones. Charming lifestyle look. Audio: welcoming French dialogue, quiet shop bell, faint espresso hiss. No subtitles, no on-screen text. (16:9, 1080p, 8s)

Why it works: The short French greeting fits an 8-second clip cleanly and lets the accent read naturally.

20. British documentary host

A seasoned host in a wax jacket stands on a windswept coastal cliff, sea crashing far below. Use @Image1 as the host's appearance. In a measured British accent he says to camera, "Out here, the tide decides everything." Medium wide shot, slow dolly-in as wind moves his hair. Overcast soft daylight, cool natural tones. Prestige nature-doc look. Audio: articulate British dialogue, gusting wind, distant surf. No subtitles, no on-screen text. (16:9, 2K, 8s)

Best for: On-camera hosting with a distinct accent; keep the line short so the wind SFX doesn't fight the voice. For more on realism, see the realistic-video guide.

Frequently Asked Questions

How do I write dialogue in a Seedance prompt?

Put the exact spoken words in double quotes and describe the delivery right next to them, like she says, warmly, "welcome back." Seedance treats quoted text as speech and lip-syncs it automatically, while the surrounding words control tone, pace, and emotion. Keep the line short and write only what you want heard — do not add stage directions inside the quotes.

How accurate is Seedance lip-sync, and which languages work?

Seedance generates phoneme-level lip-sync in a single pass, so mouth shapes track the actual sounds rather than a generic talking loop. It supports 8+ languages including English, Mandarin, Spanish, French, German, Japanese, Korean, and Portuguese. Sync is most reliable on close and medium shots with short lines; very fast or overlapping speech is harder to match.

How many words fit in an 8-second clip?

At a natural speaking pace, roughly 18 to 26 words fit comfortably in 8 seconds, which is about one to two short sentences. Crowding more text in makes the delivery rushed and hurts lip-sync. Aim for one clear line per clip, and use Seedance 2.5's longer single-clip durations if you need a full exchange.

How do I keep the same face across several dialogue clips?

Upload one clean, well-lit portrait and reference it with @Image1 as the character's appearance in every prompt. Reusing the same @Image1 keeps the face, hair, and wardrobe consistent from shot to shot so a presenter or character stays recognizable across a series. Add a short physical description as a backup so the identity holds even on wider framing.

Can Seedance handle a two-person conversation?

Yes. Name each speaker, give each their own quoted line in order, and describe who is on-screen when they talk, for example the woman on the left says "..." then the man replies "...". For a clean back-and-forth use a longer clip or time-coded segments so each line gets its own beat, and keep the two lines short so both fit and stay in sync.

How do I avoid burnt-in captions on the video?

Add no subtitles, no on-screen text to the prompt. Because Seedance sometimes renders quoted lines as captions, this instruction keeps the frame clean so you can add your own styled captions later. It does not mute the audio — the words are still spoken and lip-synced, just not printed on the picture.

Can I control the voice with an audio reference?

Yes. Reference a short recording with @Audio1 as the dialogue and voice track to steer timbre, accent, and rhythm toward a specific voice, then still write the spoken line in quotes so the model knows the words. This is useful for keeping one narrator's voice consistent across a voiceover series or matching an existing brand voice.

Advertisement