Veo 3 Vertical 9:16 Video with Dialogue: How To

To get a Veo 3 video in vertical 9:16, set the aspect ratio to 9:16 before you generate. Veo 3.1 renders vertical natively, so there is no crop. For dialogue, put the spoken line in quotes after the speaker, for example: The woman looks at the camera and says, “You’re going to want to see this.” Veo 3.1 returns the clip with voice, sound effects and ambience. On VdoBloom every Veo 3.1 clip is 8 seconds with sound: 25 credits on Veo 3.1 Fast, 100 on Veo 3.1 High Quality. There is no Google subscription and no 1:1 square option, and one clip can’t run longer than 8 seconds.

Quick facts

• Aspect ratios: 16:9 or 9:16. No 1:1.
• Length: 8 seconds per clip, fixed.
• Sound: comes back on every clip, with no toggle and no extra charge.
• Price: Veo 3.1 Fast 25 credits; Veo 3.1 High Quality 100 credits.
• Prompt: up to 5,000 characters. Auto-Translate (on by default) turns non-English prompts into English first.
• Vertical caveat: Reference Mode (three photos) is 16:9 only. Text to Video and Frame Transition work in 9:16.
• Free start: 10 signup credits, which is less than one Veo clip. One-time packs never expire.

How do I get Veo 3 videos in vertical 9:16?

Early Veo 3 only made landscape video, and many guides still tell you to crop. You don’t need to any more. Veo 3.1 supports 9:16 natively, so the model composes the frame for a phone screen instead of cutting the sides off a wide shot.

1. Open Text to Video with Veo 3.1 Fast, or pick Veo 3.1 in the model list and choose Fast or High Quality.
2. Set the aspect ratio to 9:16.
3. Keep Generation Mode on Text to Video, or use Frame Transition with a first and last frame. Both work in vertical.
4. Write a prompt framed for a tall screen (tips below) and generate.

If you start from photos, make them vertical too. In Frame Transition, the photos guide the first and last frame, so a landscape photo in a vertical job fights the frame. Reference Mode, which uses three photos to hold a product or character, renders 16:9 only. If you need that, plan a horizontal cut or switch to Frame Transition.

Prompt for the tall frame

• Put the subject in the centre third. Platform buttons cover the bottom and right edges of Shorts, Reels and TikTok.
• Frame people medium or close. “Medium close-up, waist up” fills a vertical frame. Wide shots leave a tiny figure in a lot of sky.
• Use vertical movement. Tilts, a crane up, or a subject walking towards camera suit 9:16 better than long sideways pans.
• Say it in words too. Adding “vertical framing for a phone screen” helps the model compose for the shape.

How do I make a Veo 3 video with dialogue and audio?

Veo 3.1 makes the voice, lip movement and sound in one pass. You direct it with the prompt. A clear structure works best:

Shot + speaker + how they say it + the line in quotes + sound.

Example: Medium close-up of a barista in a green apron behind a busy counter, vertical framing. She leans towards the camera and says warmly, “First coffee’s on us today.” Background: espresso machine hiss, quiet cafe chatter.

Rules that make dialogue land

• Name the speaker every time. “The older man says” beats a bare quote. With two people, describe each one so the model knows who talks.
• Keep it short. An 8-second clip fits about one or two short sentences, roughly 15–20 words. Long lines get rushed or cut off.
• One line per speaker, in order. For a back-and-forth, write: The man asks, “Ready?” The woman replies, “Always.”
• Describe the delivery. Whispers, shouts, laughs, a British accent, a slow tired voice. Tone words change the performance.
• Use straight quotes around only the spoken words. Keep directions outside the quotes, or the model may read them aloud.
• Write the line in the language you want spoken. Auto-Translate turns the rest of the prompt into English. If you need the dialogue in another language, check the result, and turn Auto-Translate off if it changes your line.

Sound effects and ambience

Ask for sound in plain words after the action: “Sound: rain on a tin roof, distant thunder.” Name two or three sounds, not ten. If you don’t want music, say “no music.” If you want silence under the speech, say “quiet room tone.”

Common failures and quick fixes

Problem	Likely cause	Fix
Clip comes back horizontal	Aspect left on 16:9, or Reference Mode used	Set 9:16 before generating; use Text to Video or Frame Transition
Subject is tiny in the frame	Wide-shot prompt in a tall frame	Ask for a medium close-up and centred subject
Words are mumbled or cut off	Line too long for 8 seconds	Cut the line to one short sentence
Wrong person speaks	Speakers not described	Describe each person and name who says each line
Stage directions are spoken aloud	Directions inside the quotes	Keep only the spoken words inside quotes
Music drowns the voice	No sound direction	Add “no music” or “soft background only”
Prompt is refused	Content rules, or a real person’s likeness	Rephrase; see when AI video ignores your prompt

What does it cost?

What you make	Veo 3.1 Fast	Veo 3.1 High Quality
One 8 s vertical clip with dialogue	25 credits	100 credits
4 takes to find the best line reading	100 credits	400 credits
A 32 s Short from four 8 s clips	100 credits	400 credits

A practical habit: draft and test lines on Fast, then render the final take on High Quality only if Fast falls short. The price is the same in 9:16 and 16:9, and failed renders are refunded. Credits are pay-as-you-go with no subscription, and one-time packs never expire. See the pricing page.

Making longer vertical videos

Each clip is 8 seconds, so a longer Short is a sequence. Write one line of dialogue per clip, keep the character’s description word-for-word the same in every prompt, and cut the clips together in your editor. For a steadier look between shots, use Frame Transition with a frame from the end of the last clip as the first frame of the next.

Where to go next

• Use Veo 3 without a Google subscription
• Veo 3.1 guide: modes and prompts
• Google Veo 3.1 review
• Best aspect ratio for YouTube Shorts
• Veo 3.1 Fast model page

FAQ

How to get Veo 3 videos in vertical 9:16?

Choose 9:16 as the aspect ratio before you generate. Veo 3.1 renders vertical natively, so nothing is cropped. On VdoBloom this works in Text to Video and Frame Transition; Reference Mode is 16:9 only.

How do I make a Veo 3 video with dialogue and audio?

Describe the shot, name the speaker, describe how they speak, and put the exact words in quotes, then add a short sound line. Keep dialogue to one or two short sentences per 8-second clip. Sound comes back on every clip at no extra cost.

Does vertical cost more than landscape?

No. Veo 3.1 Fast is 25 credits and High Quality is 100 credits per 8-second clip in either 9:16 or 16:9.

Can Veo 3 make a square 1:1 video?

Not on VdoBloom. Veo 3.1 offers 16:9 and 9:16 only. For a square clip, crop a 9:16 or 16:9 result in your editor, or use another model.

Can two characters talk in one clip?

Yes. Describe both people and write each line in order with who says it. Keep the exchange short so it fits 8 seconds.