Media tools
Sume media tools by API: trim, filter, captions, frames, inspect, speech-to-text, upscaling, background removal, transitions, and Timeline assembly.
Start with Video editing API limits: Sume media tool caps in one table

Is my Short silent? Check has_audio before detach or captions
Video inspect with frames off returns probe.has_audio. Check it before audio detach or speech captions, which refuse a clip with no audio track.

Write an instrumental ad bed prompt for Lyria 3.5 on Sume Music
A prompt pattern for a no-vocals music bed on Sume Music: 1 to 5,000 characters, no duration field, $0.125 per generation. What to put in the prompt.

Sume TTS word timestamps and sentence slices: the 70 ms boundary rule
Ask Sume TTS for timestamps.words and segmentation.mode sentence to get gapless sentence segments. The cut lands 70 ms after the last word unless you change it.

Sume TTS output: mp3, wav or raw, picked by the next step
Sume TTS defaults to mp3 at 44.1 kHz and 128 kbps; wav and raw cover editing and slicing. Pick the container, rate and encoding by what you do next.

Freeze the last 1.5 seconds of a Short for an end card
Hold the final beat of a Short as a still: pull the frame at duration minus 1.5 s with Sume video frames, then lay it as a still slot in Timeline 1.0.

Duck music under a voiceover on Sume Timeline: duck_db, gain, fade
Sume Timeline 1.0 soundtracks take gain_db, loop, fade_out_seconds up to 10 and duck_db from 0 to 20. Settings for a voiceover over a music bed.

Reel series: keep one product's look across five AI video episodes
How to keep one product looking the same across a five-episode Reel series by pinning a frame or passing references on Sume, and which models accept which.

Trim a voiceover take with Timeline audio: source_in and duration
Cut a Sume-hosted voiceover with POST /v1/timeline-1.0/audio: one part with source_in and duration, wav by default, $0.01 flat, Idempotency-Key required.

TikTok 'Reference to Video' vs Sume frame_images for product moments
TikTok's Symphony Creative Studio lets advertisers prompt exact images and products at set moments. How the job maps to Sume frame_images and references.

How do I cut a 15-second series trailer from my Shorts episodes?
Use POST /v1/video-trim with start and end on a Sume-hosted master: $0.02 per cut, so five trailer beats cost $0.10 and join in one Timeline render.

How do I caption a Shorts series in two languages in one batch?
Run one Sume caption job per episode and language: $0.20 each for clips up to 60 seconds, so a 10-episode season in two languages is $4.00.

How do I make a 180-second Shorts episode from 30-second Wan clips?
A 180-second episode is six 30-second Wan 3.0 clips joined by one Timeline render: $22.50 for video at 720p plus $0.30 for the render, in Sume prices.

How do I reuse one theme track across every Shorts episode?
Detach or build the audio once, then join parts with Timeline audio concat: $0.01 per job, up to 20 ordered parts in one gapless file.

How do I pull a Shorts thumbnail frame from an episode with Sume?
Call POST /v1/video-frames with a time in seconds to get a durable still from a Sume-hosted episode. Sume bills the job by its Modal compute.

Review 30 episode exports fast: video-inspect seek fast vs precise
Sume video-inspect gives 8 stills by default. Seek fast jumps to a keyframe at or before each time; precise lands on the frame. Pick by the check.

Duck music under dialogue and fade out a TikTok mini drama episode
Sume's Timeline render takes a soundtrack with duck_db (0-20) and a fade_out_seconds up to 5, so dialogue stays clear and each episode ends on a clean fade.

TikTok auto captions or burned-in captions on an AI video?
Use both when it matters: TikTok's auto captions are a viewer toggle that needs speech; burned-in captions from Sume ($0.20 per clip up to 60 s) always show.

Reels on the Instagram TV app: caption for a living room
Reels on a TV autoplay with sound, but a screen across the room needs bigger text. Raise font_size_ratio in a Sume caption job and test one frame first.

Reel auto-captions misspell names and brands: fix with script_text
Auto captions stumble on names and brands. Send your script as script_text so Sume keeps the speech timings and prints your spelling. Errors and the $0.20 job.

Caption a Reel after First Draft: burn once on the final cut
First Draft shifts where every word lands, so burn captions only on the final cut and restyle with source_caption_id. The $0.20 job, errors and corrections.

Logo still over a Wan 3.0 clip: Timeline compose overlay costs $0.02
Burn a GPT Image 2.5 logo or price card over a Wan 3.0 clip with Timeline compose for a flat $0.02. Import both files first; the still lasts the whole clip.

Bilingual captions crowd a vertical Reel: lay out two lines
Two caption tracks overfill a 9:16 frame. Keep cards short, centre them, and use Sume's caption design limits for words, width and height. Korean needs Hangul.

YouTube Shorts series: pin one caption look for every episode
Shorts series run from 2026-09-23. Omit style and Sume picks one from the text. Pin style, font and design in one config so ten episodes look the same.

100 Kling motion-control clips: the budget, reserve and refund
At $0.1575 a second on Sume, 100 ten-second motion-control clips reserve $157.50. A table for 5 to 30 seconds, plus how queueing, 402 and refunds behave.

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.

Kling motion control job done: a signed Python webhook receiver
Submit a Sume Kling 3.0 motion-control job with mode webhook, then verify the HMAC signature in Python, reject an empty secret and answer fast. Runnable code.

Loop a 30-second music bed under a 3-minute Reel: the body
A short bed can cover a 180-second Reel with soundtrack.loop, duck_db and a fade-out. One Timeline 1.0 body, the 3-minute bill, and the limits to respect.

MAI-Voice-2.1 is not on Sume: make talking clips with Sume TTS
Microsoft MAI-Voice-2.1 launched 2026-10-01 but Sume does not list it. Here is the path that ships: Sume TTS audio, then H3 Max lip sync on a still.

MiniMax H3 Max lip sync API: your first clip in Python
Submit a still and a Sume-hosted audio file to POST /v1/minimax/h3-max/lip-sync, poll the job, and read the result. Python stdlib, with the 5 to 14.8 s rule.

MiniMax H3 lip sync at 2K? Sume offers 480p, 768p and 1080p
MiniMax H3 the video model lists 2K, but Sume's H3 Max lip-sync route stops at 1080p. The three resolutions, their derived per-second prices and how to choose.

MiniMax H3 Max video or H3 Max lip sync: fixing the wrong_tool error
generate_video refuses minimax/h3-max/lip-sync with wrong_tool. Use avatar-image-to-video_create for lip sync, minimax-h3-max for text or image video.

Reddit's 15-second view rule: a lip-sync clip that stays under it
Reddit's Engaged Video Views beta bills videos over 15 s at 15 s. Sume H3 Max lip sync tops out at 14.8 s of audio, so one clip fits. Specs, math and a caveat.

Reel captions in two languages: language hint vs auto-detect
How Sume video captions pick a language and a style: the optional language hint, auto-detect, the Hangul rule that returns 400, and a Korean Reel example.

Shorts episode title card: a 2-second still, then the clip
Open each Shorts episode on a held title still with Timeline 1.0: slot math, the 0.5 s coverage rule, what stills ignore, and one render body you can copy.

Shorts spec sheet against Sume Timeline output: what is confirmed
YouTube lists 3 minutes, 1080p and a 100-character title for Shorts. Sume Timeline renders 1080x1920 by default. A table of what is confirmed and what to check.

Speed up a Reel? Sume has no speed field, so shorten the cut
Sume's trim, filter and timeline docs list no playback-speed control. Here is what to use to hit a Reel or Short length instead, with the exact limits.

Timeline transition_not_frame_aligned: why a 0.01 s fade fails
A Timeline fade under half a frame rounds to zero frames and is refused with transition_not_frame_aligned. How many frames a fade is at 24, 25, 30 and 60 fps.

transition_snapped_to_frame: a 0.3 s fade at 25 fps is 8 frames
Timeline rounds a fade to whole frames and reports transition_snapped_to_frame as a warning, not a failure. What 0.3 s becomes at 24, 25, 30 and 60 fps.

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.

Can a YouTube Short be square? Rendering 1080x1080 in Sume Timeline
YouTube says Shorts can be square or vertical. How to export 1080x1080 from Sume Timeline, what fit does to a 9:16 source, and the 3-minute ceiling.

Reels profile grid crops to 3:4 (1080x1440): check a frame
A Reel cover is center-cropped to 3:4 on the profile grid, per a third-party guide. Pull the frame with Sume video frames and keep faces and titles inside.

Why a 2-minute YouTube upload is or is not a Short in 2026
YouTube treats square or vertical uploads up to three minutes as Shorts if posted on or after October 15, 2024. What to check, and how to render one in Sume.

Transparent background AI image on Sume: GPT Image 2.5 field
Only the GPT Image 2.5 rows take background on Sume's image route. OpenAI says transparency needs png or webp output. A request and an alpha check in Python.

How to turn an AI image into an og:image link preview in Python
Ideogram 4.5 on Sume makes a 2:1 image at 1K as 1440x720. A Pillow crop to your link-preview size, plus the og:image:alt tag Open Graph asks for.

Can a YouTube Shorts episode run 4 minutes? Split with source_in
No: the Shorts series limit is 180 seconds. Split a long cut into two parts with Sume Timeline audio.source_in and video source_in, one render per part.

AI video upscaler API fields: factor 1.1 to 4 and fast, standard, pro
The Sume Video Upscale request takes upscale_factor from 1.1 to 4, duration_seconds from 1 to 30 and an enhancement_tier. What they do and cost.

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.

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.