Musician teaser video from cover art and a 20-second cut
Make a 20-second single-release teaser on Sume from cover art, a phone clip and a cut of your own track: one Timeline render at $0.10 for up to a minute.

A musician can make a 20-second teaser for a single from cover art, one phone clip and the best 20 seconds of the track using Sume's Timeline render. The song is the audio spine, the pictures are slots with start times, and the whole render costs $0.10 because Timeline bills $0.10 per rounded-up output minute and uses no model inference. It is assembly, not generation, so the result is exactly your art and your audio.
Use only audio you own or have cleared. A Timeline render does not clear rights for you, and platforms match uploaded audio against catalogues.
How do I pick the 20 seconds?
Timeline's audio.source_in is an in-point into a single audio file, and audio.duration_seconds is the output length. Set source_in to the start of your chorus or hook, set the duration to 20, and the render plays that section only. Add fade_out_seconds under output so the cut does not end abruptly. Avoid audio.parts here: it cannot be combined with source_in.
What does the render request look like?
Import the audio, the cover and the phone clip into your workspace first, because every URL has to be Sume-hosted. Then send something like this, with your own artifact URLs:
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: single-teaser-001" \
-d '{
"audio": {
"url": "https://media.sume.com/artifacts/artf_demo/track.wav",
"source_in": 62,
"duration_seconds": 20
},
"video": [
{ "source_url": "https://media.sume.com/artifacts/artf_demo/cover.jpg", "start": 0, "duration": 6 },
{ "source_url": "https://media.sume.com/artifacts/artf_demo/studio.mp4", "start": 6, "duration": 10,
"transition": { "type": "fade", "duration": 0.5 } },
{ "source_url": "https://media.sume.com/artifacts/artf_demo/cover.jpg", "start": 16, "duration": 4,
"transition": { "type": "dissolve", "duration": 0.5 } }
],
"output": { "fade_out_seconds": 1 }
}'What can go wrong and what does it cost?
Run the unbilled plan endpoint first. It checks the schema and the Sume-host URLs and returns the duration and the estimated cost. Output defaults to 1080 by 1920, which suits Reels, TikTok and Shorts. Video slots must start at 0 and increase, and coverage can stop at most half a second before the end of the audio.
| Item | Rate | Cost |
|---|---|---|
| Timeline render, 20 seconds | $0.10 per output minute | $0.10 |
| Ten teasers (A/B hooks) | $0.10 each | $1.00 |
| Instrumental bed from Music, if you have no track | $0.125 per audio | $0.125 |
Can Sume write a bed if I have no finished track?
Music 1.0 runs on Google Lyria 3.5 and can make a track from a text prompt, with section markers such as an intro from 0:00 to 0:30. That is useful for a lyric-video backing, but a teaser for your own single should use your own master. Try two or three different 20-second cuts with different source_in values: at $0.10 a render, testing hooks is almost free.
Sources
Related posts
More in Use cases
- Nine stickers in one render: a 3x3 transparent sheet, sliced
Ask ChatGPT Image 2.5 for a 1:1 sheet of nine stickers on a transparent background, cut it into nine PNGs, and flag any sticker that crosses a cell edge.
- Pan de muerto bakery clip in English and Spanish, Nov 1-2
Animate one photo of your pan de muerto for Nov 1-2, 2026 with Wan 3.0, then burn English and Spanish caption cues. Two versions cost $0.775 on Sume.
- Photo to cyanotype print: Qwen Image edit or a Pillow colorize?
A cyanotype look is two tones, so Pillow can do it free. Use qwen/qwen-image at $0.025 only when you want paper texture and brushed edges. Both steps here.
- Photo to felt and knitted looks: three Grok Imagine calls, 7.5 cents
Grok Imagine on Sume returns one image per call and costs $0.025, so three craft looks of one photo are three requests and $0.075. One short script runs them.
Written by Sume