Check a Short has sound and picture before adding it to a YouTube show
YouTube asks shows for high-quality video and sound and says no-visuals videos likely will not count. Probe each Short with Sume video inspect first.

How do you check a Short before it joins a show?
Run it through a probe and a handful of stills. YouTube's shows page lists high-quality video and sound as a quality standard and says videos with no visuals likely will not be considered a show, so each episode needs both a picture and a sound track.
Sume's video inspect reads one media.sume.com clip, never re-encodes it, and returns probe facts and stills. The default is a synchronous call that waits up to 30 seconds and returns 200, or returns 202 and you poll.
| Field | Content |
|---|---|
| probe | Facts about the file, including has_audio |
| frames[] | Stills with t, url, width and height |
| transcript | Only with transcribe true, $0.01 per audio minute |
| warnings[] | Soft notes |
The three checks
First, sound. A probe with frames set to false is enough to read has_audio. If you ask for a transcript on a clip with no audio track, the API refuses with inspect_source_has_no_audio, so probe before you transcribe.
Second, picture. Leave frames unset for 8 mid-bin stills and look at them: if they are the same image, the episode is a static hold. Third, shape. Each still carries its width and height, so portrait or square is visible at a glance.
- has_audio false: fix the mix.
- Eight near-identical stills: add picture slots.
- Width greater than height: re-render before conversion.
- Duration over 180 seconds: split or trim.
Using it across a season
Import each file first, since the API does not fetch the open internet, and send an Idempotency-Key on every inspect so a retry is safe. The stills are durable media.sume.com images, so you can reuse the best one as a thumbnail base.
Transcripts help in one more way: reading the transcript catches a wrong word before it goes live, at one cent per audio minute. Inspect itself is billed by its compute, so read the rate in the catalog rather than assuming it.
What it does not do
Inspect returns facts, not a verdict. It cannot tell you whether YouTube will treat a show as a show, because the help page gives no pass mark. It only confirms that each file has the basics.
Keep the answers in your ledger next to each episode, with the job id of the inspect call. When a render is replaced, run the same three checks again, so the show never holds an episode that skipped them.
Sources
Related posts
More in Media tools
- compose_video_has_no_audio: a silent clip under a still on Sume
Timeline compose warns compose_video_has_no_audio and still completes. The output is silent. How to add a voice-over afterward with a Timeline render.
- Copy space in an AI image: leave room for text, overlay it in code
Ask for empty space in the prompt, check the region with Pillow, and put the headline on top in code, so the text is exact and the picture stays a picture.
- Custom thumbnail for each Short in a YouTube series: pull the frame
YouTube lets a Shorts series carry a unique custom thumbnail per Short. Pull the best frame from each render with Sume video frames, then dress it.
- Estimate a video render cost before you pay: Sume timeline plan
POST /v1/timeline-1.0/plan returns billable minutes, an estimated cost and a filtergraph summary without a job, a reserve or a media download.
Written by Sume