Google Vids caption font, position, animation: Sume design overrides

Google Vids captions can now be customized for font and typography, position on screen, and animation, according to a Google Workspace Updates post dated 30 September 2026. Sume's standalone caption job maps to two of those three directly: placement and motion are design overrides. Typography is partly covered, and a free choice of Latin font is not available, because the font field only takes Hangul faces.

How do Google's three controls map to Sume fields?

Can I pick a brand font in Sume captions?

Not for Latin text. The docs list font as an optional Hangul face, and naming one beside a Latin style returns a 400 (caption_font_requires_hangul_style). The Latin styles slam, punch and tiktok-green draw in their own display faces. Google's post says the point of the update is matching captions to brand fonts and colors; in Sume you can match colours and weights, not the family.

Also note that design is not supported on punch or tiktok-green, so use slam (or a Hangul style for Korean speech) when you want overrides.

What does a request with overrides look like?

Each field is optional and merges over the style's own value, so one key changes one thing. Numbers outside their documented range come back as a 400 instead of rendering wrong, so check the response before batching. The video_url must be a fetchable public HTTPS URL, and a media.sume.com clip qualifies.

curl -X POST https://api.sume.com/v1/video-captions \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: caption-design-001" \
  -d '{
    "video_url": "https://media.sume.com/artifacts/artf_demo/promo.mp4",
    "style": "slam",
    "language": "en",
    "design": {
      "colors": { "active": "#22D3EE" },
      "placement": { "anchor_ratio": 0.75 },
      "phrasing": { "max_words": 3 },
      "motion": { "enter_seconds": 0.15 }
    }
  }'

How do I try a second look without paying for a second transcript?

Pass source_caption_id instead of video_url. Sume reuses the first caption's source video and word timings, so no second speech-to-text runs. Billing is unchanged, because a restyle is still a render: the docs list $0.20 per accepted standalone caption job for videos up to 60 seconds, under the current fixed estimate. Confirm the live figure in GET /v1/catalog.