SYSTEM NOTICE

Auto translation by AI. Be sure, accuracy, nuances and authorial intent may not be fully reflected.
見出し画像

Create immersive videos with music and dialogue using Veo 3 in the Gemini app

Gemini - Google の AI

Hello, this is the official note editorial department for Google's AI, "Gemini".

Did you know that Gemini's "video generation feature" has been updated?

Google AI Pro plan and Google AI Ultra plan users, as well as those on the Google Workspace Enterprise plan, can now easily generate videos using Google's latest video generation model, Veo 3!

* The Google AI Pro plan offers access to Veo 3 Fast, which is optimized for high quality and speed, while the Google AI Ultra plan provides access to Veo 3, which can generate videos at the highest quality level. For Google Workspace plans, you can maximize your use of AI features and models by using Google AI Ultra for Business as an add-on. Please see here for details.

With Veo 3, your ideas are transformed into wonderful videos complete with music and dialogue. You can turn your free-flowing imagination directly into video, such as "a video of a cat becoming an astronaut and taking a walk on the moon" or "an animation of a dolphin illustration coming to life."

By being able to turn ideas into videos, you can, for example, give an original short movie for a friend's birthday, or express daily events and feelings of gratitude in a one-of-a-kind video.

Communication with family and friends is sure to become even more fun and enriched.

In this article, we will introduce how to use the Gemini Veo 3 video generation feature, its characteristics, and specific examples of how it can be used in work and daily life!

* Explanations regarding the use of generative AI, including Gemini, are for illustrative purposes. Please verify the accuracy of actual response results yourself. Please use generated videos only for image verification purposes.

Click here for the Veo 2 article.


How to use the Gemini video generation feature "Veo 3"

It is very easy to use. Simply select "Video" in the prompt input field at the bottom of the Gemini screen and enter the image of the video you want to generate as text.

Once you enter the text, your idea will become a video with sound in 2 to 3 minutes.
If you refine your prompt, you can even make the characters in the video speak.

Features of the Gemini video generation feature "Veo 3"

With Veo 3, we have leveled up from Veo 2, and it is now possible to generate not only videos but also music and dialogue at the same time.

By entering the image of the video you want to generate as text, you can generate an 8-second video. The video is in landscape orientation, and the current resolution is 720p.
Here are 5 features of Veo 3.

1. Native audio generation and lip-syncing

You can generate not only visuals from text but also synchronized audio (dialogue, sound effects, music). When making a character speak, it can automatically generate audio with natural intonation and emotion, and the movements will be natural, including not only lip movements but also subtle changes in facial expressions that match the content being spoken.

② Improved prompt understanding and narrative capability

Understanding of complex prompts has been improved, allowing for the depiction of consistent stories and characters. You can visualize emotions and situations, such as "a cat looking up at the sky sadly in the rain," and have characters appear in different scenes. This enables the creation of not just standalone clips, but short films with a clear beginning, middle, and end, as well as continuous narratives.

③ Realism and physics simulation

You can express realistic footage, including light, shadows, textures, and fluid motion. The videos capture realism so convincing you might forget it's CGI, such as light scattering on a water surface, trees swaying gracefully in response to wind strength, and fabric draping according to gravity.

Create a video of green tea being poured from a black, textured teapot into a teacup.
Depict the poured tea filling up the cup.
Make the sound of the tea being poured feel crisp and vivid.


④ Support for diverse styles and high resolution

You can generate videos in a wide variety of styles, from live-action to anime-inspired or specific vintage aesthetics. You have the freedom to create claymation, watercolor, or cyberpunk visuals. It meets all needs, from short videos for social media to creating materials for full-scale video works intended for large-screen projection.

Generate a video in the style of miniature clay art. A man's hands are slightly visible, slicing bonito tataki with a knife in a kitchen. The background should feature small soy sauce and vinegar bottles as cute accents.

⑤ Image to Video

Why not add motion to your own photos or your child's illustrations with Veo 3?

For example, a landscape photo taken on a trip can be transformed into a video that captures the atmosphere of the moment, with leaves rustling in the wind and the water surface shimmering.

If it's an illustration drawn by your child, you can make it come alive by having the character's hair blow in the wind, making their eyes blink, or even having them spin around.

It's very simple to use. Just upload an image and enter a simple instruction into Veo 3, and you can create a video without any professional editing skills.


A memorable photo might tell a new story as a dynamic short movie.

6. Integration with Google Flow

You can integrate with "Flow," an AI filmmaking toolkit that enables more advanced editing and camera work control. You can seamlessly connect multiple generated video clips on a timeline, or fine-tune camera angles and movements later. This enables video production that combines high-speed AI generation with precise editing and direction by human creators.

Videos you can generate with Veo 3 and use cases

Next, we will introduce the situations where Veo 3 can be utilized.
It can be used in various scenes, from daily life and private use to work and learning.

Make your new campaign project presentations more attractive

It is difficult to fully convey the passion and worldview put into a project with just words or still images. Veo 3 is useful for grabbing the listener's attention in the first 30 seconds and making your presentation memorable.

For example, you can easily create a video expressing Google AI Pro with "summer fireworks display" using Veo 3.

In addition, you can visualize abstract concepts such as "a new life starting at a school building with cherry blossoms dancing" or "an EV speeding through a near-future city." Since you can share the project's intent and target image with your team or clients, it will help you convey your ideas accurately and make your presentation more attractive.

Liven up the venue at an event opening

Veo 3 is also useful when you want to surprise participants and liven up the venue with a new initiative, such as at a company event or club opening. You can even easily create an original cat musical video to make the start of the rainy season fun.

A scene from a masterpiece musical film. Amidst a downpour of rain hitting the city, many cute cats holding colorful umbrellas are dancing happily in unison. They energetically stomp in puddles, creating magnificent splashes. In the center, the lead cat wearing a bright yellow raincoat performs a dynamic dance with a big smile. The cats sing loudly, "I love the rain!", and their singing voices create a beautiful harmony with the sound of the rain. The bright daylight reflects off the raindrops, making the whole world sparkle.

Additionally, for creating a company kickoff movie, simply input "a grand CG animation of employees flying into space, themed around this term's slogan 'Challenge the Future'" to share your vision and goals visually, not just in words. If you weave successful case studies and the mission of each department into the video, it might also increase the sense of unity as an organization.

Add entertainment to communication with friends and family

Veo 3 is also useful when you want to turn your SNS or message replies into more interesting reactions and surprise your friends.

You can express yourself after finishing a week of housework and work as a Shiba Inu, or express a message to friends and family wishing them sweet dreams through a unicorn video; you can deliver feelings that cannot be conveyed by words alone through video.

A Shiba Inu wearing a suit and sleeping soundly and happily on a bus


A unicorn running through the sky with sweets, in a fairy-tale anime style. It says "have a nice dream" in sync with a "twinkle" sound effect.

You can also use unique short videos to express your feelings and situations to others, such as sending a "celebratory video of colorful fireworks going off in the night sky" to a friend's happy news, or conveying your feelings with a "video of cute animals giving a standing ovation with heartfelt gratitude".

With just one prompt, you can create a one-of-a-kind reaction that matches the vibe of the moment.

Turn your child's imagination into a video

You can make playtime with your children and bedtime stories even more special with Veo 3.

For example, try telling them a story you came up with together, like "a brave hero meets mysterious creatures and crosses a rainbow bridge." The story will come to life like magic.

The experience of not just listening to a story, but "creating" it together, will also become a lasting memory.


It's also recommended to try animating illustrations drawn by your children or yourself.

Prompt tips for mastering Veo 3

Veo 3 allows you to easily create videos by entering your ideas as "prompts" (instructions).
Here are some tips to take your videos to the next level, such as how to customize the background music or make characters speak dialogue.

① Clarify the "characteristics" of the characters: appearance, emotions, actions, voice, etc.

Write down the characteristics, personality, and actions of the characters, people, or animals specifically. By delving deep into not just their physical appearance, but also their individuality, habits, and emotional state, their presence will stand out.

Example: "A detective wearing a worn-out trench coat," "a curious fox cub," "a cheerful tin robot," etc.

Also, "sound"—such as dialogue, animal cries, or mechanical noises—is an important element in creating a world or scene. Be sure to clearly specify the tone and characteristics of the voice as well.

Example: "Whisper '...' in a low, raspy voice," "Bark 'Woof!' happily," "Speak 'Mission accomplished' in an electronic voice," etc.

* As of August 2025, the only officially supported language is English.

② Build the world of your video: atmosphere, music, and sound effects

To draw viewers into the world of your story, it is important to carefully design the mood, music, and sound effects of the footage. By adding sensory details, you can provide a deeper immersive experience.

Try incorporating the genre of music, the instruments used, and the tempo into your prompt to match the emotions of the scene you want to generate (joy, sadness, tension, grandeur, etc.). By doing so, you can generate mood-setting background music and immersive sound effects that are perfect for the scene.

Example: "A lighthearted orchestral fanfare signaling the start of an adventure," "The sound of a character's footsteps splashing in a puddle, the sound of the wind, the sound of church bells ringing in the distance," etc.

③ Define the visual expression: visual style and camera work

To determine the final look of your video, it is important to combine visual styles and camera work. Choose the style that best fits the world of your video, such as animation, live-action, claymation, cyberpunk, or pixel art.

Example: "Warm, hand-drawn animation like a picture book," "Cyberpunk live-action movie style," "Comical clay art of a clumsy penguin head chef slipping on milk in the kitchen and causing a scene," etc.

Camera work is the act of designing the "perspective" of a story. Therefore, incorporating it into your prompt will enhance immersion in the video. For example, camera work that looks up at a subject from a low angle can express awe or the size of the subject, while shaky handheld camera movement can convey a character's frustration or a tense situation.

You can even express the perspective of an animal, which you don't usually get to experience. Why not manipulate the perspective and create an immersive video?

Example: "A slow zoom-in on a mysterious ancient book placed on a dusty table," "A dramatic low-angle shot capturing a giant robot towering over the city," "A wide pan shot from left to right revealing a futuristic cityscape at night," etc.

The more specific and detailed your instructions are, the easier it will be to accurately capture the image in your mind and reproduce it as video, so please give it a try!

Gemini's video generation will continue to evolve

With Gemini's Veo 2, I was able to turn the images in my mind into concrete videos without needing professional video editing skills.

The video generation Veo 3 introduced this time has evolved even further, not only generating footage but also automatically generating background music, sound effects, and dialogue that harmonize with the scene.

You can now shape the images in your mind into immersive videos filled with movement and sound, allowing for even more immersive video generation.

You don't need professional video editing or music knowledge to use Veo 3. By simply entering a short prompt, you can express the nuances of emotion and the atmosphere of the video that words alone cannot convey, both visually and aurally.

Add a little movement and sound to your usual messages. That alone will make your interactions more fun. The Google AI Pro plan is available for a one-month free trial, so please try out the new video generation!