Inspect old Sora mp4s before mixing them with new clips
Saved Sora files may not match new clips in size, length or frame rate. Probe them with Sume video inspect, read the facts, then decide what needs re-rendering.

Before you cut saved Sora files together with new Sume clips, probe each file. Sume's video inspect returns probe facts, sampled stills and optional transcript for one clip, without re-encoding it. It reads clips on media.sume.com owned by your workspace, so import the file first. The probe tells you duration, size and codec facts; the stills let you eyeball cuts. Then decide which old shots to keep and which to regenerate on a current model.
What the endpoint needs
The docs set a few firm rules that shape the workflow.
| Rule | Detail |
|---|---|
| Source | One media.sume.com clip owned by the workspace; no open-internet fetch |
| Import first | POST /v1/media-imports |
| Idempotency-Key | Required |
| Default mode | sync, waits up to 30 seconds; 202 if still queued |
| Default stills | 8 mid-bin stills, or 1 fps when under 8 seconds |
| Output | Probe, frames as media.sume.com images, optional transcript, warnings |
A short audit loop
For a folder of saved Sora exports: import each file, call inspect with frames: { at: [0, 2] }-style explicit timestamps or the default, and store the probe record next to the file name. Group by what the probe reports, then compare with the model you plan to use for new shots. Mismatches in aspect ratio or duration are the common reason a mixed edit looks wrong.
Decide per clip
A clip that matches your target can go straight into an edit. A clip that is close may need a trim or reframe with the other media tools. A clip that is wrong in a way tools cannot fix gets re-rendered. Because a re-render on a current model is a different clip, not the same shot, check the stills against the new output before approving the swap.
- Keep the original file; probe results are records, not replacements.
- Use
fastseek for skimming andprecisewhen a timestamp must match. - Do not transcribe unless you need the audio text.
Honest limits
Inspect does not judge quality or content semantically; the docs describe it as probe facts, stills and optional speech-to-text. It bills the probe by compute, so check your usage after a large audit. For the wider job lifecycle, see the jobs guide.
Sources
Related posts
More in Media tools
- LinkedIn video thumbnail under 2 MB: pull the frame with Sume
LinkedIn video ad thumbnails are JPG or PNG up to 2 MB. Pull a still from your clip with Sume's video frames endpoint and check its size before upload.
- LTX-2.5 48 fps option: conform frame rate on Sume
LTX-2.5 offers 24/25 fps or 48/50 fps. Sume's video catalog has no fps field, but Timeline output.fps conforms a clip to 24, 25, 30 or 60.
- Lyria 3.5 output: MP3 or WAV at 44.1 kHz, and what Sume returns
Google lists MP3 by default or WAV, 44.1 kHz stereo, with a SynthID watermark. Sume's music docs say the artifact is typically audio/mpeg. Check it in code.
- Lyria 3.5 prompts: [Verse] tags or [0:00-0:30] time ranges?
Google's Lyria 3.5 docs show [Verse], [Chorus] and [Bridge] tags. Sume's music docs show time-range markers. What each says, and how to test both on one brief.
Written by Sume