Podcast audio to a 45-second teaser video with Timeline source_in

Timeline 1.0 takes an audio spine with a source_in in-point and a still as a static hold, so a podcast excerpt plus cover art renders as a video with no editor.

4 min readSume
All posts

A podcast teaser video needs three things: an excerpt of the audio, one image to look at, and a video file to post. Timeline 1.0 does all three in one render. audio.source_in sets the in-point into a single audio file, audio.duration_seconds sets the length, and a still in video[] is held on screen for the whole slot. No waveform or caption is drawn; the picture is static.

Which fields make the excerpt?

Timeline 1.0 fields used for an audio-led teaser, from the Timeline 1.0 docs, read 2026-10-01.
FieldEffect
audio.urlOne Sume-hosted spine. Exclusive with audio.parts.
audio.source_inIn-point into that single file. Output length is still duration_seconds.
audio.duration_secondsOutput length, 1 to 1800 seconds.
video[].source_urlA Sume-hosted clip or still. A still is a static hold.
output.width / heightEven integers 256 to 2160. Default output is 1080x1920.

How do I find the 45 seconds worth posting?

Run video inspect on a video version of the episode with transcribe: true and sentence segmentation to get timed text. The transcript is billed at the speech-to-text rate, $0.01 per audio minute on the docs page, and the probe and stills are unbilled. Pick the sentence that opens the excerpt and use its start time as source_in. Video inspect needs a video file, so an audio-only file needs a separate speech-to-text call.

What does the render request look like?

This one takes a 45-second excerpt starting 12 minutes 30 seconds in and holds one cover still for the whole length at 1080x1080. Both URLs must already be this workspace's media.sume.com files. Send mode: "sync" if you want the finished job back in the response; the default is async.

curl -X POST https://api.sume.com/v1/timeline-1.0/render \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: podcast-teaser-001" \
  -d '{
    "audio": {
      "url": "https://media.sume.com/artifacts/artf_demo/episode.wav",
      "source_in": 750,
      "duration_seconds": 45
    },
    "output": { "width": 1080, "height": 1080 },
    "video": [
      { "source_url": "https://media.sume.com/artifacts/artf_demo/cover.png",
        "start": 0, "duration": 45 }
    ]
  }'

What does it cost, and can I add a fade?

The render is billed at $0.10 per output minute, and the docs round the output up to a whole minute, so a 45-second teaser counts as one. Add output.fade_in_seconds and fade_out_seconds (0 to 5 each) to soften the edges of the excerpt. To check the document without billing, call POST /v1/timeline-1.0/plan first.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume