YouTube podcast RSS makes static-image videos: add a real trailer

When you submit a podcast RSS feed, YouTube builds a static-image video from your show art for each episode you choose. If you also record video, a short captioned trailer gives the channel something that moves. Sume's video trim and video captions make that clip; the upload is your step.

YouTube's side is from Deliver podcasts using an RSS feed, read 2026-10-03.

What does YouTube do with an RSS feed?

The help page lists these behaviors:

• It creates a static-image video from the podcast's show art for each episode and uploads it to your channel.
• Selected episodes first arrive as private videos.
• Content uploaded this way cannot contain advertisements; branded content and sponsorships must be declared.
• Once published, the audio cannot be updated. A fix means a new video, and the old one is set to private.
• RSS ingestion is available in select countries and regions.

Why add a trailer?

The RSS route gives you the episode as a still image over audio. It does not give you a clip built for a feed. A 30 to 60 second excerpt with burned-in captions works muted and can be uploaded as its own video, separate from the RSS-created episode. Nothing on YouTube's page says how such clips perform, so treat it as a test.

How do I cut the clip?

Import the recorded episode to media.sume.com with POST /v1/media-imports, then trim. Video trim takes a start and exactly one of end or duration (0.2 to 900 seconds), from sources up to 1800 seconds, at $0.02 per job. An end past the source clamps and returns a trim_clamped_to_source warning.

curl -X POST https://api.sume.com/v1/video-trim \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: podcast-trailer-001" \
  -d '{
    "video_url": "https://media.sume.com/artifacts/artf_demo/episode.mp4",
    "start": 312.5,
    "duration": 45,
    "precision": "exact"
  }'

How do I caption it, and what about sponsors?

Video captions transcribes the clip, or skips transcription if you pass cues or segments with text, start and end. A clip with no speech fails as caption_no_speech. It costs $0.20 for clips up to 60 seconds.

Pick an excerpt with no host-read sponsor, since YouTube's page says RSS content cannot contain ads and asks you to declare sponsorships. Sume does not detect ads or judge policy; that review is yours.

What does Sume not do here?

Sume does not submit RSS feeds, edit the show-art video YouTube makes, or change audio after publishing. Read finished files from GET /v1/jobs/:id/result (Jobs and results).