Talking Avatars

Talking Avatars turns a photo and a voice into a lip-synced talking video. Type what they say in any voice, or upload the audio.

Talking videos are made on the Image & Video page, in the composer's Lip sync mode.

Make a talking video

Open Image & Video in the sidebar, and choose Lip sync at the bottom left of the composer. Or choose a face in Characters on the Explore tab.
Choose Add a photo and pick a photo. Or choose Make it talk on an image you made.
Under What they say, choose one:
- A script: type the words in the box, choose a Voice, and choose Hear it to preview. The row shows "about N s".
- A recording: choose Choose audio and pick the file.
Choose the Look: Monologue or Anchor.
For Anchor, choose a Shape. For Monologue with a script, you can describe the style of movement in the line above the box, for example "calm and friendly, small hand gestures".
Tick the confirmation: "I have the right to use this person's photo and voice, and won't use the video to impersonate, mislead or harm anyone."
Choose the arrow button.

With a script, Cinara makes the voice first (Making the voice…). Talking videos take one to three minutes. You can leave the page and come back.

Monologue: Natural gestures, most affordable. The video follows the photo's shape.
Anchor: Steadier head, faster to make. Aspect ratio 16:9, 9:16, 1:1, 4:5 or 5:4.

Voice (script only): your workspace's cloned and designed voices, then Cinara's voices.
The settings row shows the estimated length, "about N s".

Photo: PNG, JPEG or WebP, up to 20 MB. A clear, front-facing photo of one face works best: a person, or a drawn or animal character.
Script: up to 900 characters.
Speech: under 60 seconds. The page shows "Keep it under 60 seconds." when it's too long.
Uploaded audio: MP3, WAV or M4A, up to 20 MB.
The confirmation box must be ticked.
Cinara doesn't make talking videos of well-known people.
If a photo can't be used, you see "This photo couldn't be animated. Use a clear, front-facing photo of one face."

Monologue: 562 credits per second of video (shown as 5,620 per 10 s).
Anchor: 1,000 credits per second of video (shown as 10,000 per 10 s).
The video's credits are rounded up to the nearest 10.
Type a script also costs Speech credits: 1 credit per character.
The composer shows about how many credits, charged for the length of the video.
Monologue videos can run a little past the speech, so about 1.5 seconds extra is reserved. You're charged for the video's real length, never more than was reserved.
If the video fails, your credits are refunded.