Synthesia API ratios include 4:5 and 5:4; Sume avatar video does not

Synthesia's create-video reference lists aspectRatio values of 16:9 (the default), 9:16, 1:1, 4:5 and 5:4. Sume's avatar talking-video route accepts 1:1, 3:4, 9:16, 4:3 and 16:9, with 9:16 as the default. The overlap is 16:9, 9:16 and 1:1. If your feed placement is 4:5 (the portrait feed shape), Synthesia produces it directly and Sume's closest native choice is 3:4, a slightly taller frame.

Sources: Synthesia's create-video reference, read 2026-10-04, and Sume's avatar video guide.

Side by side

Resolution matters too. Sume's resolution is currently 720p for avatar video; this post does not state Synthesia's output sizes because the reference page read for it does not give them.

Choosing 3:4 when the placement wants 4:5

3:4 is 0.75 wide-to-tall and 4:5 is 0.8, so a 3:4 clip is a little taller than a 4:5 slot. Platforms that crop to fit will trim the top and bottom of the 3:4 clip. That is fine when you keep the face and product in the centre band, and risky when you burn text near the edges.

• Keep the face between 15 and 85 percent of the frame height.
• Put captions in the middle third, not at the very bottom.
• Preview a still first, see the preview guide, before paying for a full render.

Fixing the shape after the render

If you must have exactly 4:5, Sume documents a video filter route among its post-processing models; test a crop on a short clip first. Cropping a 3:4 clip down to 4:5 loses a little height, which is the safe direction.

Do not stretch the clip. A stretched face is the first thing viewers notice.

Bottom line

Pick the vendor by ratio only if 4:5 or 5:4 is a hard requirement. Otherwise 9:16, 1:1 and 16:9 cover most placements on both. Keep a short table like the one above in your brief so the editor and the engineer agree before the first render.