Edit Video Soon

Model

OmniHuman v2

Photo + audio to a lip-synced, expressive video

Details

What it does: Image + audio → lip-synced video. Emotion and movement follow the audio, for speech, song or dialogue.
Price: 1 credit per second of video (billed on audio length, minimum 4 credits). Failed runs are refunded.
Image: JPG, PNG, WebP, BMP, TIFF, GIF, HEIC or HEIF. Up to 30 MB, aspect ratio 0.4–2.5, 300–6000 px per side.
Audio: WAV or MP3, up to 15 MB, 4–120 seconds.
Prompt languages: Chinese, English, Japanese, Korean, Spanish (Mexican) and Indonesian.

Take a look around. Sign in to upload files and generate.

A clear photo of one face. JPG, PNG, WebP, BMP, TIFF, GIF, HEIC or HEIF, up to 30 MB. Between 300 and 6000 px per side, aspect ratio 0.4–2.5.

Upload photo

MP3 or WAV, 4–120 s, up to 15 MB.

Upload audio

No audio? Generate a voice from text

1 credit per 100 characters · max 500 characters

Supported languages: Chinese, English, Japanese, Korean, Spanish (Mexican) and Indonesian. Prompts are screened against our Acceptable Use Policy.

One photo. One voice. One video.

Real samples made with OmniHuman v2. Your own videos will show up in History as soon as you generate them. See pricing