Skip to content

Use case

Pull the teaching moments out of a full session.

A session recording holds a handful of explanations that stand on their own. Upload it once and they come back as vertical 9:16 clips with the captions burned in.

Source

Upload the session recording.

MP4, MOV or WebM, within your plan's size and length ceilings, in one file. A session past your plan's length has to be split before it goes up.

The upload panel, with a source file staged and the run settings on the right.
Transcript

The whole session gets timed.

Audio is split into five-minute chunks, several are transcribed at once, and the results are stitched back into one timeline with word-level and segment-level timings. A two-hour session is not four times the wait of a 30-minute one.

The same workspace while a run is still cutting.
Candidates

Explanations come back as ranges.

The model returns every strong qualifying candidate up to your plan's limit, ordered by score and carrying its ranges, title, description and hashtags. Ranges can sit far apart in a long session; a clip can join up to four of them.

One finished clip as it appears in the list.
Output

Vertical clips that keep the slide.

Clips render vertical at 9:16, 1080 x 1920 or 720 x 1280. The default framing scales the whole source frame to the width and pads above and below with a blurred copy of it, so a shared screen is not cropped away. Fill-frame is the other option and crops to the centre instead. While the source is retained, either can be swapped on the finished clip and re-rendered.

The clips workspace, rendered by the app's own components with a sample run.

Sample run

What one Q3 session returned.

A single 68-minute recording, put through one run. Three explanations came back scored, one of them stitched from two moments that sat half an hour apart. These are that run's own numbers.

Source file
webinar-migrations-q3.mp4
Length
68:26
Credits charged
69
Clips returned
3
Ranges cut
4 across 3 clips
Highest score
8.4
That clip's title
Run both schemas for a fortnight before you delete anything
Start clipping
The clips workspace, rendered by the app's own components with a sample run.

The parts that matter here.

Transcription
Reframe
Export

What a run costs.

A session is metered by its own running time, rounded up to the whole minute. The 68:26 recording above came to 69 credits; two clips or five out of it would have cost the same, because the charge is on the file and not on the output.

See pricing

Other recordings this works on.

Podcasts

Turn a 60-minute episode into a queue of reviewable vertical cuts.

Interviews

Find the compact answer without replaying the full recording.

Creator content

Get a week of short posts out of one long upload.

Business content

Cut the recordings your team already has into short clips.

Session questions.

Your plan sets the ceiling for a single file, and a longer upload is refused before anything is charged. Split the recording and run the parts separately, or move to a plan with a longer ceiling.

With the default framing, yes. The whole source frame is scaled to the full width and the space above and below is filled with a blurred copy of it, so nothing at the edges is lost. Fill-frame crops into the centre and can cut a slide in half; while the source is retained, it can be swapped for the pad and re-rendered without spending credits.

No. Only the audio is transcribed, and the model reads that transcript. Text on a slide is never extracted and has no influence on which moments come back; it is simply part of the picture that gets reframed.

On the plans that include branding, yes. A logo image is drawn above centre and a handle line under the captions, on every clip in the run. Both are set before the run starts, so a whole session comes out marked the same way.

MP4, MOV and WebM. The size and length ceilings are set by your plan, from 2 GB and 60 minutes at the lowest to 8 GB and four hours at the highest. The pricing page lists both per plan. A file past your plan's limit is rejected before anything is charged.

Ready when the recording is

The session already taught it once. Post it again.