Image & Video

Image & Video is one page for stills, film shots and talking presenters. Open Image & Video in the sidebar. (Old links to Images, Video and Talking Avatars open this page too.)

The page has two tabs, Explore and History, and a composer at the bottom of the window where you make everything.

Explore

Explore is a gallery of images and videos made in Cinara.

• Hover over a video to play it silently.
• Choose any image or video to see it large with its prompt. Use this prompt puts the prompt into the composer, switched to Image or Video to match.
• Characters: a row of faces above the gallery. Choose one to make it talk: it becomes the photo in Lip sync. Your photo switches to Lip sync so you can add your own.
• Narrow the gallery with the bar that stays at the top while you scroll: search, and All, Images, Videos or Lip sync.

History

Everything this workspace has made here, newest first: images, videos and lip-synced videos. The same search and filters work. Choose one to open its result; Back to Image & Video returns.

The composer

Choose what to make at the bottom left of the composer: Image, Video or Lip sync. Type into the box, check the settings on the row below it, and choose the arrow button (or press ⌘ Enter, Ctrl Enter on Windows). The price shows beside the button.

Image

• Describe the picture, for example "A lighthouse on a cliff at sunset, painted in oil".
• Image refs: add pictures to edit instead of starting from nothing, up to 4. Several pictures of the same character keep it consistent. Use as reference on a result adds that image here.
• Look: the model (see Images), each with its credits per image.
• Shape: the aspect ratio.
• How many: 1 to 4 images.
• Random: the seed. Choose it to type a number; the same seed and words give you the same shot again.

Video

• Describe the shot, or with a start frame, describe the motion.
• Start frame: a still to bring to life. Animate on an image result puts it here. End frame appears for models that support one, once there's a start frame.
• Look: Scene, Feature or Director's Cut (see Video).
• Shape (without a start frame), Size (resolution, where the model offers it), Length, and the seed (where the model supports it).

Lip sync

• Add a photo: a clear, front-facing photo of one face, or choose a character in Explore. Make it talk on an image result puts it here.
• What they say: A script (type the words in the box, pick a Voice, and choose Hear it to preview; the row shows "about N s") or A recording (choose Choose audio).
• Look: Monologue or Anchor (see Talking Avatars). Anchor adds Shape. With Monologue and a script, a line above the box takes the style of movement, for example "calm and friendly, small hand gestures".
• Tick the confirmation under the box: "I have the right to use this person's photo and voice, and won't use the video to impersonate, mislead or harm anyone."

Results

• Images: each image has Animate, Make it talk, Use as reference, an Editor button and a download button. Edit settings loads the prompt and settings back into the composer; Generate again makes new images.
• Videos: play, Edit (opens it in the Editor) and Download; Edit settings and Generate again.
• Lip sync: play, Edit and Download; Generate again.

Credits, limits and model details are in Images, Video and Talking Avatars.