Ovi Audio-Video Generator

Ovi is Character.AI’s open video+audio generator (paper with Yale; GitHub character-ai/Ovi) — not a clipping editor and not a TTS-only desk. It makes picture and sound together from text, or from text plus a first frame. Public model card: 5-second or 10-second clips (Ovi 1.1 adds the 10-second window) at 24 fps, 960×960, with 9:16, 16:9, 1:1 and other ratios listed. Speech tags drive dialogue; an Audio: line describes effects and music. Architecture: 5B visual branch on a WAN 2.2 5B backbone, 5B audio, 1B fusion. Hosted playgrounds (for example WaveSpeed) may list a 5-second 540p row — treat that as a hosted SKU, not an ImgToVid generate row. Skip dollar prices. ImgToVid does not run Ovi; the generator on this page drafts on Seedance 2.5.

01 — Capabilities

A 5/10s AV generator, not a TTS studio or a long-to-short cutter

Public Ovi pages sell a short clip with native sound. They do not sell Noiz WAV/MP3, or OpusClip’s 0–15 minute cut of footage you already have.

01
5 or 10 seconds at 24 fps, 960×960
Ovi 1.1: 10 seconds; original card also 5 seconds. Ratios include 9:16, 16:9, 1:1. That table is Ovi’s, not Veo’s 4 / 6 / 8 seconds.
02
Video and audio in one pass
T2AV or I2AV (still + prompt). Speech tags and an Audio: description. Lip-sync is learned, not a boxed face tracker.
03
Hosted WaveSpeed rows stay hosted SKUs
Some hosts list 5 seconds at 540p. Treat those as their table, not an Ovi generate row on this site.
04
Not Noiz TTS, not OpusClip minutes
Noiz is a voice studio. OpusClip cuts a long video into shorts. Ovi generates 5 or 10 seconds of picture with sound. This site does not open Ovi.

02 — Prompt to output

Prompts that match the clips above

Each prompt describes the picture on its card. “Use this prompt” sends that text to Create. The generator on this page runs Seedance 2.5, not Ovi.

01 · Portrait
PromptClose-up of a young man with curly hair in a grey knit sweater, soft window light. He glances away, then smiles. Slow push-in, shallow depth of field.
02 · Product
PromptA white leather sneaker on a pale grey studio backdrop, soft diffused light. The camera orbits slowly. Stitching stays sharp, the look stays minimal.
03 · Street
PromptA slow dolly down a cobblestone lane in an old European village. Climbing roses on stone walls, café tables in morning sun, a few people walking in the distance.
04 · Character
PromptA young woman with shoulder-length brown hair in an olive t-shirt, late-afternoon gold light. She turns her head and looks off. Breeze in her hair, a handheld feel.

03 — Where it fits

Jobs an AV generator is selling

Public pages point at a 5 or 10 second take with speech and effects. They do not sell a Noiz voice file or an OpusClip short of Ovi’s own.

A 5 or 10 second talking take from text or a still
T2AV / I2AV on the public card. This page does not run that weight.
Dialogue plus sound effects in the same pass
Speech tags and Audio: lines. That run stays on Character.AI’s stack or a host you pick.
A WAV from a script, or a 0–15 minute clip of existing footage
That is Noiz or OpusClip. Ovi is the 5/10s AV model.
A draft shot on ImgToVid
Write the shot here. Seedance 2.5 carries the prompt into Create. Membership and credits are checked there.

04 — Verdict

What Ovi actually is

Strengths

Character.AI Ovi: video+audio from text or a still; OSS weights
Public card: 5 or 10 seconds (Ovi 1.1), 24 fps, 960×960, 9:16 / 16:9 / 1:1
Speech tags + Audio: line; 5B visual / 5B audio / 1B fusion
Honest that 5/10s is Ovi’s table, not a Veo 4/6/8s generate row here

Trade-offs

No Veo-style 4 / 6 / 8 second picker of Ovi’s own — it is 5 or 10 seconds
Hosted 540p rows are not this site’s Ovi generate table
Ovi is not available on ImgToVid
Marked coming soon on this site

05 — Side by side

Ovi, Veo 3.1 and Kling 4.0

Ovi is a short AV generator. Veo 3.1 and Kling 4.0 numbers below are from those vendors’ public docs. Ovi’s 5 / 10 second card is not an Ovi generate row on ImgToVid.

	Ovi	Veo 3.1	Kling 4.0
Developer	Character.AI	Google DeepMind	Kuaishou
Single clip	5 or 10 seconds at 24 fps (Ovi 1.1); hosted desks may list 5s	4, 6 or 8 seconds	3–30 seconds
Picture	960×960; 9:16 / 16:9 / 1:1 listed; some hosts list 540p	720p, 1080p, 4K at 24 fps	720p, 1080p and 4K listed
Sound	Native with the picture (speech tags + Audio: line)	Native, with the picture	Stereo, with lip sync
Start from	Prompt, or a still plus prompt	Prompt, still, first and last frame	Keyframes or 15 references
On ImgToVid	Coming soon	Coming soon	Coming soon

06 — How it works

Start a clip on ImgToVid

Step 1
Describe one shot
Subject, action, camera. Ovi will not be the model that runs it.
Step 2
Use the generator on this page
It runs Seedance 2.5, because Ovi is not offered here. The prompt and the settings carry into Create.
Step 3
Generate in Create
Membership and credits are checked there. Coming back to this page does not switch the model over to Ovi.

07 — FAQ

Ovi, in plain terms

What is Ovi?

Ovi is Character.AI’s video+audio model: text or still in, 5 or 10 seconds of picture with sound out. OSS on GitHub. ImgToVid does not host Ovi.

How long is an Ovi clip?

Public card: 5 seconds or 10 seconds (Ovi 1.1) at 24 fps. That is not a Veo 4 / 6 / 8 second picker. Hosted desks may list 5 seconds only.

Does Ovi generate video from text?

Yes, T2AV, plus I2AV from a still. Sound is generated with the picture. That is Character.AI’s model (or a host), not a cinematic one-shot on ImgToVid.

What picture size can I use?

960×960 on the public card, with 9:16 / 16:9 / 1:1 listed. Some hosts list 540p. This site does not list an Ovi generate-row resolution.

Can I generate Ovi videos on ImgToVid?

No. The generator on this page runs Seedance 2.5 so you can draft a shot. Ovi is marked coming soon.

Can I use the results commercially?

Commercial use of anything you generate on ImgToVid depends on your plan. Check Pricing and our Terms before you publish, and make sure prompts and uploads do not infringe someone else’s rights.