AI Dialogue Video Generator

Build a conversation out of single-speaker clips. Each video takes one photo and one audio track, so record one side of the dialogue, generate that person, then do the same for the other speaker and edit the cuts together.

Input: One speaker photo + that speaker's audio · Output: MP4, 4–120 seconds

Start generating See pricing

Sample: Podcast host · Host photo + voice clip. Generated with OmniHuman 2.

How it works

Step 1
Upload a photo of the first speaker: yourself, someone who consented, or an original fictional character.
Step 2
Add only that speaker's audio for this beat of the conversation.
Step 3
Generate the clip, repeat for the next speaker, and assemble the dialogue in your editor.

Why use it

One face per clip

The model does not stage two people in one shot. Separate clips keep each performance tied to the right photo.

Speech, not a script engine

You supply the spoken audio. The tool does not write the conversation or assign roles.

Pauses read on the face

Breaths, laughs and hesitations in the track show up in the performance.

Short exchanges

Each video is 4 to 120 seconds long. A longer scene is a sequence of clips.

Example direction prompts

Optional text you can add in the studio to guide the scene, movement and camera.

“Person at a cafe table, listening posture breaking into a reply, natural eye contact.”
“Interview guest leaning forward, thoughtful pause, then a direct answer.”
“Friend on a couch, casual disagreement, small hand lift, relaxed face.”
“Podcast guest in a quiet room, headphones off, nodding once before speaking.”

Frequently asked questions

Can two people appear in one video?

No. Each generation is one photo and one audio track, so it is one speaker. Create a clip per speaker and cut between them.

Can I upload a recording of both voices?

You can, but one face will perform the whole track, including the other person's lines. For a real exchange, split the audio so each clip contains one speaker.

Can I use a celebrity interview?

No. Impersonating real people, including public figures, is prohibited. Use people who consented or original fictional characters, and only audio you have rights to.

Does dialogue use a different mode?

No. Dialogue uses the same photo-plus-audio input as speech. There is no separate conversation mode.

What does a clip cost?

1 credit equals 1 second. A 15-second line uses 15 credits. Failed generations are refunded.

Make your first video

Plans start at $9.99/month. 1 credit = 1 second of video, and failed generations are refunded.

Open the studio Pricing