Alibaba text to video with sound and dialogue
WAN 2.7 AI Video Generator
WAN 2.7 turns a written shot list into a finished 1080p scene with sound and spoken dialogue, all from a single prompt.
We made every clip on this page with WAN 2.7 on Pixel Dojo: a three shot sci fi scene, a neon motorcycle ride in the rain, a storm radio call, a glacier collapse, a 15 second night market food spot and a claymation kitchen. Each one comes with the exact prompt behind it, so you can start from it and make it yours.
10 credits ($0.63) per 5 s clip at 720p. Included in every plan, from $10/month.
Already a member? Sign in
By Brian Lovett · Updated
What WAN 2.7 makes best
WAN 2.7 is Alibaba's AI video generator with the soundtrack built in. On Pixel Dojo you describe a scene and get back a 720p or 1080p clip from 2 to 15 seconds long, with an audio track and spoken dialogue made together with the picture.
Its standout skill is multi shot storytelling. Write Shot 1, Shot 2 and Shot 3 into one prompt and WAN 2.7 cuts them into a single scene, like our astronaut sequence at the top of this page. Put a line of dialogue inside a shot and it is spoken in that shot.
We wrote every prompt on this page the way a cinematographer writes a shot note: shot size and camera first, then one specific subject, one action, the light, the lens and a closing line of sound. Five aspect ratios (16:9, 9:16, 1:1, 4:3 and 3:4) let the same idea become a trailer, a vertical social post or a square product clip, and video extend mode continues a clip you already have.
What you can create with WAN 2.7
Every clip here was made with WAN 2.7, soundtrack built in, and every tile shows the prompt we wrote, word for word. Copy one, swap the subject, and keep the shot, light and sound lines.
A short film scene from one prompt: a wide shot on the dunes, a close-up inside the visor where she speaks her line, then a low angle run under two moons. Start from this prompt to write your own three shot scene.
Cinematic three shot sequence. Shot 1: Wide shot, a lone astronaut in a scuffed white suit with a gold visor trudges across rust red dunes toward a crashed silver lander at dusk, dust streaming off the ridges. Shot 2: Close-up inside the helmet, the astronaut, a woman in her forties with close cropped gray hair, stares as a green light blinks on the lander. [Commander Reyes, breathless]: "It's still transmitting." Shot 3: Low angle, she breaks into a run toward the lander under two pale moons, boots kicking up red dust. 35mm anamorphic lens, teal and amber grade. Sound: wind hiss, crunching footsteps, a rising synth drone.
Night action in 1080p. Magenta and cyan spray fans off the front tire for the whole ride, with sharp neon reflections across the flooded street. A strong start for a music video or an action trailer.
Low tracking shot at wheel height, the camera racing alongside. A rider in a scuffed red helmet and black waxed jacket leans a matte black cafe racer motorcycle through a flooded neon night street, the front tire slicing a tall fan of spray that glows magenta and cyan, rain streaking past the lens. 35mm anamorphic lens, shallow depth of field, teal and orange grade. Sound: engine roar, the hiss of water, rain drumming on metal.
A vertical 9:16 performance made for social feeds: two full turns of crimson silk under a single hard spotlight, the fabric swinging like real cloth through the haze.
Symmetrical wide shot, slow dolly in. A dancer in her thirties with a sleek black bun spins on a bare black stage in a flowing crimson silk gown, the skirt fanning into a wide arc, under a single hard overhead spotlight that cuts a cone through drifting haze, dust sparkling in the beam. High contrast, deep shadows, 35mm lens. Sound: the whisper of silk, bare feet brushing the floor, a low solo cello.
Native dialogue from a single prompt. A gray bearded man in a yellow oilskin shouts the scripted radio call from the lighthouse tower in warm lantern light while the storm rages behind the windows. Write your line as [Name, tone]: "line" and WAN 2.7 voices it.
Medium close-up, slow push in. A weathered lighthouse keeper in his sixties with a salt gray beard and a yellow oilskin coat grips a crackling radio handset in a lamp lit tower room, lightning flashing across his face as rain streams down the window behind him. He presses the button and leans in. [Lighthouse Keeper, hoarse and urgent]: "Northern Star, turn east now." 50mm lens, warm tungsten light against cold blue lightning. Sound: howling wind, rain lashing the glass, radio static. No background music.
Nature documentary scale at sunrise: a slab of blue ice crashes into a turquoise lagoon, then the camera pulls back to reveal the floating icebergs. Start from this prompt for your own landscape epic.
Aerial drone shot gliding fast and low over a turquoise glacial lagoon toward a towering blue ice cliff as a huge slab calves off and crashes into the water, throwing up a white wave and a wall of spray, small icebergs rocking in the swell, low golden sunrise light raking across the ice. 24mm wide lens, epic scale, crisp detail. Sound: a thunderous crack of ice, a roaring splash, wind.
A 15 second food spot, the longest clip WAN 2.7 makes: noodles lifted from the pot, glossy broth ladled in, a soft egg and scallions on top, and the bowl served across the counter in warm lantern light. Built like a restaurant commercial.
Continuous 15 second steadicam take, no cuts. The camera follows a ramen cook in her fifties with a navy headscarf as she lifts a nest of noodles from a steaming pot, snaps the strainer twice, slides the noodles into a bowl, ladles glossy broth, lays a soft egg and scallions on top and pushes the bowl across the counter toward the lens, steam curling through warm paper lantern light in a busy night market alley. 35mm lens, shallow depth of field, warm amber grade. Sound: bubbling broth, the clack of the strainer, a sizzling grill, distant chatter. No background music.
Claymation with stop motion charm: a badger chef flips a buttery pancake high out of a copper pan, lands it back in the pan and grins with pride in his round windowed burrow kitchen. Start from this prompt for a playful brand short.
Claymation stop motion, locked off medium shot. A round clay badger chef in a tiny white toque flips a golden pancake high out of a copper pan in a cozy cluttered burrow kitchen, the pancake spins in the air and lands back in the pan with a floppy slap, and the badger beams with pride. Warm tungsten light, visible thumbprints in the clay, felt and cardboard props, charming stop motion cadence. Sound: sizzling butter, the slap of the pancake, a cheerful ukulele tune.
Where WAN 2.7 shines: detail, faces and sound
Two more clips from our session show how far one idea can go: the rider prompt from the showcase at 720p, and a still from our astronaut scene brought back to life through the image to video API with a brand new line of dialogue.
The same rider prompt at 720p, 10 credits for 5 seconds, with the magenta and cyan spray fanning off the front tire.
Low tracking shot at wheel height, the camera racing alongside. A rider in a scuffed red helmet and black waxed jacket leans a matte black cafe racer motorcycle through a flooded neon night street, the front tire slicing a tall fan of spray that glows magenta and cyan, rain streaking past the lens. 35mm anamorphic lens, shallow depth of field, teal and orange grade. Sound: engine roar, the hiss of water, rain drumming on metal.
Image to video through the API (wan-2.7-i2v) from a still of our hero clip. Same face, a new line of dialogue, spoken on cue.
Slow push in on the helmet. She draws a sharp breath, fog blooms on the inside of the visor, the green light pulses across her cheek, and she glances toward the horizon. [Commander Reyes, quietly]: "Someone is still out here." Sound: steady breathing inside the helmet, wind outside, a soft electronic beep.
- Multi shot scenes from one prompt. Name each shot and WAN 2.7 edits them together, placing each line of dialogue in the shot where you wrote it.
- Native audio on every clip. Every clip we made has its soundtrack built in, and every line of dialogue we wrote was spoken as written.
- Two resolutions from the same prompt. 720p at 2 credits per second is made for social feeds, and 1080p at 3 credits per second brings out sharper neon, spray and helmet detail, as the two rider clips show.
- Up to 15 seconds in one clip, long enough for a complete food spot like our ramen stall.
- Your character keeps her face. We sent a still of our astronaut to image to video with a new line, and the same woman speaks it.
- Specific details come out as written. Wardrobe, props and lighting from our prompts, from a scuffed red helmet to a navy headscarf and a single hard spotlight, all showed up on screen.
WAN 2.7 vs WAN 2.6: one prompt, two marble lions
We sent the same prompt to both models at the same 720p and 5 second settings. WAN 2.6 is the release before 2.7. WAN 2.7 gives you the full transformation: the whole statue splits apart on its plinth and a lion with a glowing golden mane roars under the moon, framed like a title reveal. WAN 2.6 plays it as a tight close-up on the mane. On Pixel Dojo, WAN 2.7 also adds video extend with last frame control.
Same prompt: Medium wide shot, slow push in. A cracked marble lion statue in a moonlit stone courtyard shudders, its stone skin splitting and falling away in chunks to reveal a living lion with a glowing golden mane that shakes off the dust and roars, shards and dust swirling. Cold blue moonlight against the warm gold glow of the mane, 35mm lens. Sound: stone cracking, rumbling debris, a deep roar.
WAN 2.7 keeps the whole lion on the plinth as the stone peels away to a glowing golden mane under the full moon.
Medium wide shot, slow push in. A cracked marble lion statue in a moonlit stone courtyard shudders, its stone skin splitting and falling away in chunks to reveal a living lion with a glowing golden mane that shakes off the dust and roars, shards and dust swirling. Cold blue moonlight against the warm gold glow of the mane, 35mm lens. Sound: stone cracking, rumbling debris, a deep roar.
WAN 2.6 moves in close as the statue bursts apart, with a fiery mane, heavy debris and a big roar.
Medium wide shot, slow push in. A cracked marble lion statue in a moonlit stone courtyard shudders, its stone skin splitting and falling away in chunks to reveal a living lion with a glowing golden mane that shakes off the dust and roars, shards and dust swirling. Cold blue moonlight against the warm gold glow of the mane, 35mm lens. Sound: stone cracking, rumbling debris, a deep roar.
| WAN 2.7 | WAN 2.6 | |
|---|---|---|
| Credits for 5 seconds at 720p | 10 credits | 10 credits |
| The reveal | The whole statue splits apart and the living lion stands on the plinth | Slabs break away as the camera closes in |
| Framing | Whole body on the plinth, pillars and moon in frame | Moves in close on the head and mane |
| Look | Glowing golden mane and a full body roar under the moon | Fiery mane in a tight close-up |
| Modes on Pixel Dojo | Text to video, video extend with last frame, image to video through the API | Text and image to video through the API |
How to prompt WAN 2.7 for clips like these
Name your shots and it edits for you
Write Shot 1, Shot 2 and Shot 3 into one prompt and WAN 2.7 cuts them into one scene. Our 10 second astronaut sequence cuts to the close-up exactly where the dialogue sits.
Write dialogue as [Name, tone]: "line"
Every line we wrote this way was spoken on cue. Keep each line short and give the speaker a name, not a pronoun.
End with a Sound: line
List the sounds you want last, like wind hiss, crunching footsteps or a solo cello. Add No background music when you only want effects and voice.
Spend your words on subject and light
WAN 2.7 follows specific details closely. A precise wardrobe note and one clear light source, like our single hard spotlight on the dancer, go a long way toward the look you pictured.
Pick the format for where it will play
Choose 9:16 for vertical social clips like the dancer, 16:9 for trailers and scenes, and 1:1 for square feeds. Switch to 1080p when fine detail matters most.
Continue a scene through the image to video API
Send a still to /api/v1/models/wan-2.7-i2v/run with a new line of dialogue. We did this with a frame from our hero clip and got the same face speaking a new line, for 10 credits.
WAN 2.7 pricing: credits per clip
WAN 2.7 is included in every Pixel Dojo plan, and plans start at $10/month. Each clip costs credits based on its length and resolution: 2 credits per second at 720p and 3 credits per second at 1080p.
At the settings on this page, a 5 second 720p clip is 10 credits, a 5 second 1080p clip is 15 credits, the 10 second 1080p astronaut scene is 30 credits and the 15 second 720p ramen spot is 30 credits. Image to video through the wan-2.7-i2v API is 2 credits per second at 720p, so a 5 second clip is 10 credits.
You see the exact price for your settings before you generate.
WAN 2.7 questions
What is WAN 2.7?
WAN 2.7 is Alibaba's AI video generator. It turns a text prompt into a 720p or 1080p clip of 2 to 15 seconds with sound and spoken dialogue, and it can edit several named shots into one scene. You can use it on Pixel Dojo in the browser or through the API.
How do I use WAN 2.7 on Pixel Dojo?
Open the WAN 2.7 tool, write your prompt, pick a duration, resolution and aspect ratio, and generate. The fastest start is any prompt on this page: copy it, swap the subject and keep the shot, light and sound lines. Developers can send the same fields to /api/v1/models/wan-2.7-t2v/run.
Does WAN 2.7 generate sound and dialogue?
Yes. Every clip we made has its own soundtrack, built in. Write dialogue as [Name, tone]: "line" inside the shot where it should be spoken, and end the prompt with a Sound: line for effects and ambience.
How long can a WAN 2.7 video be, and at what resolution?
Each clip runs 2 to 15 seconds at 720p or 1080p, in 16:9, 9:16, 1:1, 4:3 or 3:4. In 16:9 that is 1280 by 720 or full HD 1920 by 1080. Video extend mode continues a source clip of up to 10 seconds.
How much does WAN 2.7 cost?
WAN 2.7 is included in every plan, from $10/month. A clip costs 2 credits per second at 720p and 3 credits per second at 1080p, so a 5 second 720p clip is 10 credits and a 10 second 1080p scene is 30 credits.
WAN 2.7 vs WAN 2.6: which should I use?
Pick WAN 2.7 for new projects. It is the newer release, it cuts named shots into one scene with dialogue voiced where you wrote it, and on Pixel Dojo it adds video extend with last frame control. Both cost 10 credits for a 5 second 720p clip. In our marble lion comparison WAN 2.7 framed the whole transformation on the plinth, while WAN 2.6 moved in close on the mane.
Does WAN 2.7 do image to video?
Yes, through the API. Send an image_url and your prompt to /api/v1/models/wan-2.7-i2v/run. We sent a still from our astronaut scene with a new line of dialogue and got the same face speaking it, for 10 credits at 720p.
Make your first WAN 2.7 clip
Write a prompt above or start from any example on this page. One subscription covers every model on Pixel Dojo.
Start with WAN 2.7