Making a video clip from a written description only.
What the AI does
The AI reads your description and creates the scenes, motion and look from scratch.
When to use it
When you have an idea but no starting picture.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, choose the Text → Video mode and describe the clip.
Common mistakes
Describing several unrelated scenes for one short clip.
Leaving out who or what is moving.
Example
A red fox trotting through fresh snow in a quiet pine forest at sunrise, slow tracking shot, soft warm light, calm mood.
image-to-video
Also written: i2v, image to video, animate a photo, Image → Video
Making a video clip that starts from a picture you provide.
What the AI does
The AI keeps the look of your image and adds motion and camera movement to it.
When to use it
When you need the video to match a specific photo, product or character.
Where it works
Most video generators, including Sora, Veo and Runway, offer an image-to-video option. In AI Compass, choose the Image → Video mode, pick an image, then describe the motion.
Common mistakes
Describing a different scene from the one in the picture.
Using a blurry or tiny image.
Example
Gently animate this photo: the leaves sway in the wind, the water ripples, and the camera slowly pushes in. Keep the colors and composition the same.
pan
Also written: camera pan, panning shot, pan left, pan right
The camera turns left or right from a fixed spot, like turning your head.
What the AI does
The AI sweeps the view sideways across the scene.
When to use it
To reveal a wide place or follow something moving sideways.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Mixing up pan (turning) with tracking (moving alongside).
Asking for a fast pan in a very short clip.
Example
Slow pan from left to right across a misty mountain valley at dawn, revealing a small village by a river.
tilt
Also written: tilt up, tilt down, camera tilt
The camera points up or down from a fixed spot, like nodding your head.
What the AI does
The AI moves the view vertically, for example from feet to face or from ground to sky.
When to use it
To reveal height, such as a tall building, tree or person.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Writing "tilt" without saying up or down.
Confusing tilt with a dutch angle, which is a sideways lean.
Example
Slow tilt up from the base of a giant redwood tree to its top against a bright blue sky.
dolly
Also written: dolly in, dolly out, push in, pull out, dolly shot
The whole camera moves forward or backward, as if on a small cart.
What the AI does
The AI moves the viewpoint closer to or further from the subject, so depth changes naturally.
When to use it
To build tension, focus on a detail, or slowly reveal the surroundings.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Using "zoom" when you want the 3D feeling of moving closer.
Asking for dolly in and dolly out in one short clip.
Example
Slow dolly in toward a steaming cup of coffee on a wooden table by a rainy window, cozy morning light.
zoom
Also written: zoom in, zoom out, crash zoom
The lens makes the subject look bigger or smaller while the camera stays still.
What the AI does
The AI enlarges or shrinks the view without changing the camera's position.
When to use it
For a quick emphasis or a classic, flatter zoom look.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Expecting zoom to look like walking closer (that is a dolly).
Zooming too fast so the result looks shaky or blurry.
Example
Slow zoom in on an old clock tower as the hands move toward midnight, cinematic lighting, light fog.
tracking shot
Also written: tracking, follow shot, camera follows
The camera moves along with a moving subject, keeping it in view.
What the AI does
The AI keeps the subject in frame while the background moves past.
When to use it
For running, walking, driving or anything in motion.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Not saying what the camera should follow.
Combining tracking with many other camera moves at once.
Example
Tracking shot following a cyclist riding along a coastal road at golden hour, ocean on the right, smooth camera.
orbit shot
Also written: orbit, 360 shot, arc shot, camera circles
The camera circles around the subject.
What the AI does
The AI moves around the subject so you see it from several sides.
When to use it
To show a product, a statue or a dramatic hero moment.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Asking for a full 360 turn in a very short clip.
Using a busy background that distracts from the subject.
Example
Slow orbit shot around a sleek white sneaker on a pedestal, clean studio background, soft reflections.
crane shot
Also written: crane up, crane down, jib shot, rising shot
The camera rises or lowers smoothly, as if on a crane arm.
What the AI does
The AI lifts or lowers the viewpoint, often revealing a wider scene.
When to use it
For grand openings or endings that reveal a location.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Confusing it with tilt, where the camera only turns.
Forgetting to describe what is revealed.
Example
Crane shot rising from a street musician playing violin to reveal a busy city square at dusk.
handheld
Also written: handheld camera, shaky cam, documentary style
A camera look with small natural shakes, as if a person holds it.
What the AI does
The AI adds slight wobble for a real, in-the-moment feel.
When to use it
For documentary, vlog or tense, realistic scenes.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Using it for product shots that should look smooth.
Asking for heavy shake, which can look broken.
Example
Handheld camera following a chef plating a dish in a busy restaurant kitchen, natural light, documentary style.
aerial shot
Also written: drone shot, bird's-eye video, flyover
A view from high above, like a drone flying over the scene.
What the AI does
The AI shows the scene from the sky, often moving forward or circling.
When to use it
For landscapes, cities, coastlines and travel videos.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Adding close-up details that cannot be seen from high up.
Not saying the direction the camera flies.
Example
Aerial drone shot flying forward over turquoise water toward a small tropical island, bright midday sun.
static shot
Also written: locked-off shot, still camera, fixed camera, tripod shot
The camera does not move at all; only the scene moves.
What the AI does
The AI keeps the frame steady and focuses on motion inside it.
When to use it
For calm scenes, talking characters or showing clear actions.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Also asking for camera moves in the same prompt.
Choosing a scene where nothing moves, which can look like a still image.
Example
Static shot of a candle flickering on a windowsill while snow falls outside at night.
slow motion
Also written: slow-mo, slowmo, slowed down
Action shown slower than real life.
What the AI does
The AI stretches the motion so details like splashes or hair movement are easy to see.
When to use it
For dramatic moments, sports, water, dust or fabric.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Using slow motion for a scene with very little movement.
Packing too many actions into a slow-motion clip.
Example
Slow motion close-up of a strawberry dropping into a glass of milk, splash frozen in the air, bright studio light.
time-lapse
Also written: timelapse, sped up, hyperlapse
A long period squeezed into a few seconds, so time looks fast.
What the AI does
The AI shows changes like clouds racing, a day turning to night or a flower opening.
When to use it
For sky, city traffic, plants growing or construction.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Confusing it with fast camera movement.
Not saying what should change over time.
Example
Time-lapse of clouds rolling over a mountain lake as day turns into a starry night.
loop
Also written: seamless loop, looping video, perfect loop
A clip whose end matches its start, so it can repeat smoothly.
What the AI does
The AI tries to make the last moment flow back into the first one.
When to use it
For backgrounds, social posts and repeating animations.
Where it works
Some video tools have a loop setting; others only follow the word in the prompt, and results are not always seamless. In AI Compass, write it in your video description.
Common mistakes
Expecting a perfect loop every time.
Using big camera moves, which make looping harder.
Example
Seamless loop of gentle ocean waves rolling onto a sandy beach, static camera, soft evening light.
duration
Also written: length, seconds, clip length
How long the video clip lasts.
What the AI does
The AI plans the action to fit the chosen time.
When to use it
Always: match the amount of action to the length.
Where it works
Most video generators make short clips and let you pick the length in their settings. In AI Compass, pick the duration in Options; clips are a few seconds long.
Common mistakes
Describing a whole story for a few-second clip.
Writing a length in the prompt that differs from the setting.
Example
A short clip of a paper airplane gliding across a sunny classroom and landing on a desk.
fps
Also written: frames per second, frame rate, 24fps, 60fps
How many still images are shown each second of video.
What the AI does
Higher numbers give smoother motion; around 24 gives a film-like feel.
When to use it
Mostly for editing and export choices.
Where it works
Some tools let you choose the frame rate; many pick it for you. AI Compass chooses the frame rate itself, so you do not need to write it.
Common mistakes
Writing "60fps" and expecting a generator to follow it.
Mixing up frame rate with resolution.
Example
Smooth, film-like shot of an old red train passing through golden autumn hills, cinematic look, steady camera, soft afternoon light.
resolution (720p / 1080p)
Also written: 720p, 1080p, HD, Full HD, resolution
How many pixels the video has; 1080p is sharper than 720p.
What the AI does
The AI renders the clip at the chosen size; higher resolution shows more detail.
When to use it
Pick higher resolution for big screens and lower for quick tests.
Where it works
Most video tools set resolution in their settings, not in the prompt. In AI Compass, choose 720p or 1080p in Options; 1080p is available only for the longest clip.
Common mistakes
Writing "4K" in the prompt and expecting a 4K file.
Using the highest setting for every test.
Example
A crisp, detailed shot of a hummingbird hovering near a red flower in a garden, sunny day.
vertical video (9:16)
Also written: 9:16, portrait video, vertical, reels format
A tall video shape made for phones, used by Reels, TikTok and Shorts.
What the AI does
The AI frames the scene for a tall screen, keeping the subject in the middle.
When to use it
For social media stories and short videos.
Where it works
Many video tools let you choose the shape in their settings. AI Compass makes vertical 9:16 clips.
Common mistakes
Describing wide landscapes that lose their sides in a tall frame.
Placing key action at the edges.
Example
Vertical video of a barista pouring latte art into a cup, subject centered, warm café light.
video quality (Normal / HD / Professional)
Also written: Normal, HD, Professional, quality tier
AI Compass's video quality choice: Normal, HD or Professional.
What the AI does
Higher tiers use stronger video models for more detail and smoother motion, and cost more credits.
When to use it
Use Normal to test ideas and a higher tier for the final clip.
Where it works
This is an AI Compass setting chosen on the video screen, not a prompt word. Other tools have their own quality or model options.
Common mistakes
Using the top tier for every draft.
Expecting quality settings to fix a vague description.
Example
A close-up of rain drops falling on a puddle in a city street at night, neon reflections, slow motion.
sound (audio)
Also written: audio, sound effects, music, with sound
The sounds in the clip, such as ambience, effects or music.
What the AI does
Video models that make audio try to match sounds to what happens on screen.
When to use it
When the clip should feel alive, e.g. waves, rain or a crowd.
Where it works
Some video generators can create sound and some cannot. AI Compass clips include sound; describe the sounds you want in words.
Common mistakes
Expecting exact songs or copyrighted music.
Asking for clear spoken lines in a very short clip.
Example
Waves crashing against rocks at a windy coast, seagulls calling, deep ocean sounds, gray stormy sky.
shot list (scenes)
Also written: scenes, scene list, storyboard, shots
A plan of the shots or scenes in the order they should appear.
What the AI does
The AI follows the order of shots you describe, as far as the clip length allows.
When to use it
When you plan a longer video made of several clips.
Where it works
Video generators read scenes as plain words; short clips usually fit only one or two shots. In AI Compass, write it in your video description.
Common mistakes
Listing many scenes for one short clip.
Changing characters' looks between shots.
Example
Shot 1: a hiker reaches a mountain ridge at sunrise. Shot 2: wide view of the valley below glowing in golden light.
hook
Also written: opening hook, attention grabber, first seconds
The first seconds that grab attention so people keep watching.
What the AI does
The AI puts the most surprising or interesting moment at the start.
When to use it
For social videos and ads where viewers scroll fast.
Where it works
This is a plain-language idea that works in any video or script tool. In AI Compass, write it in your video description.
Common mistakes
Starting slowly with a long intro.
Making a hook that has nothing to do with the rest.
Example
Open with a close-up of a glass shattering in slow motion, then reveal it is a new unbreakable bottle bouncing on the floor.
video styles (Cinematic, Comedy, Drama...)
Also written: Auto, Cinematic, Adventure, Action, Comedy, Thriller, Drama, Product / Ad, Social Media
The overall style or genre of the clip, like cinematic, comedy or thriller.
What the AI does
The AI adjusts pacing, lighting, framing and mood to fit the style.
When to use it
When you want a clear feeling, not just a scene.
Where it works
Video generators take the style as plain words. AI Compass shows Style options (Auto, Cinematic, Adventure, Action, Comedy, Thriller, Drama, Product / Ad, Social Media); also write the style in your description so it is clearly applied.
Common mistakes
Picking a style but describing the opposite mood.
Mixing several genres in one short clip.
Example
Cinematic thriller style: a lone figure walks down a foggy alley at night, slow tracking shot, cold blue light, tense mood.
product ad video
Also written: product video, commercial, ad, Product / Ad
A short clip that shows off a product in an attractive way.
What the AI does
The AI focuses on the product with clean backgrounds, smooth moves and good lighting.
When to use it
For online shops, social ads and launches.
Where it works
Works as plain words in any video generator; image-to-video helps keep your real product's look. In AI Compass, write it in your video description.
Common mistakes
Expecting brand logos or text to appear perfectly.
Using a busy background that hides the product.
Example
Product ad: a glass perfume bottle rotating slowly on a marble surface, soft studio lighting, gentle reflections, elegant mood.
camera speed
Also written: slow camera, fast camera, smooth movement, gentle motion
How fast the camera moves: slow, smooth or fast.
What the AI does
The AI matches the camera's speed to your words, which changes the mood.
When to use it
Slow for calm or luxury, fast for action and energy.
Where it works
Video generators such as Sora, Veo and Runway take this as plain words in the prompt. In AI Compass, write it in your video description.
Common mistakes
Not giving any speed, so motion looks random.
Asking for very fast moves, which can cause warping.
Example
Very slow, smooth dolly in toward a sleeping cat on a sunny windowsill, calm and peaceful mood.
transitions
Also written: cut, fade, crossfade, match cut
How one shot changes into the next, like a cut or a fade.
What the AI does
Generators may follow a described change, but transitions are usually added later in editing.
When to use it
When joining several clips into one video.
Where it works
Most AI video tools make single clips; transitions are added in a video editor. In AI Compass, describe one shot per clip and join clips in an editor.
Common mistakes
Asking for many transitions inside one short clip.
Expecting exact editing effects from a prompt.
Example
Smooth fade from a city street at sunset into the same street glowing with lights at night, static camera.
character consistency
Also written: same character, consistent character, keep the same face
Keeping the same person or character looking the same across clips.
What the AI does
The AI tries to keep face, clothes and style the same when you describe them the same way or give a reference image.
When to use it
For stories and series with recurring characters.
Where it works
Many video tools do better with a reference image (image-to-video); results still vary. In AI Compass, use Image → Video with the same picture, or repeat the exact same description.
Common mistakes
Describing the character differently each time.
Expecting perfect matches between separate clips.
Example
A young woman with short red hair, a yellow raincoat and green boots walks through a rainy park, medium shot, soft gray light.
Every guide and dictionary card is also in the AI Compass app, with saved items, search and “Try it” buttons that open chat, image or video with the example filled in. Reading is free; creating uses credits, and the price is shown before you start.