Automated video production pipeline

Raw footage in.
Final video out.

Upload your footage. Get a video with voiceover, captions, and SEO metadata. All bundled in one pipeline.

one job, end to endpipeline-demo-60s.mp40:58

Same pipeline, three caption presets.

Pick a preset per job and it is burned into the render: classic_bold for long-form, minimal_clean for demos and course work, highlight_brand for short-form. Nobody opens a subtitle editor.

classic_bold

Classic Bold

Large white text, heavy outline, center-bottom.

Warehouse module release notesource18:22finished1:5216:9
minimal_clean

Minimal Clean

Smaller text, softer outline, barely-there shadow.

Q3 changelog walkthroughsource12:47finished1:3816:9
highlight_brand

Highlight Brand

Largest text, golden accent colour, uppercase forced.

Pricing page teardownsource08:11finished0:589:16

Customer story, recorded on location

2:0716:9customer story

Onsite: port terminal field day

3:2416:9field report

Onboarding: first invoice run

2:3616:9how-to

Open any one of them to see the source file, the template it ran under, every stage it passed through, and what it cost.

One file in. A horizontal, a vertical, and a silent loop out.

one source file

Interview setup under available light

customer-interview-a-cam.mp4

duration
24:05
size
7.6 GB
scenes found
12

Scene detection runs once. Each finished video below reuses those scenes under a different template, so none of this work is repeated.

Customer story

The full arc: the problem, the switch, the number they quote at the end.

runtime
2:07
aspect
16:9
template
customer story
scenes kept
12 of 12
minutes charged
2.1 min

Different footage, different length, same finished shape.

template in force

Release note

opening card
2s, title from the script
voice
narrator 04, neutral
captions
burned in, 2 lines
runtime ceiling
2:00
closing card
3s, product wordmark

Each job copies this template at creation. Editing it later changes the next job, never one already running or open at review.

input: screen recording

Warehouse module release note

source
18:22
scenes
9
finished
1:52
input: talking head

Why we moved reconciliation nightly

source
09:44
scenes
6
finished
1:47
input: event capture

What shipped, told from the floor

source
41:38
scenes
17
finished
1:58

Autopilot when you are moving fast. Manual control when it matters.

The script is written from your own footage. You never touch a keyboard to get one. Leave review off and the job runs straight through to the finished file.

Switch it on for the videos where the wording matters and the pipeline stops here instead. Approve it, edit it, or leave it open. Nothing downstream runs and nothing is charged until you answer.

awaiting script reviewopen3d 04:12obc-7f39-4ab

script / draft 1 / 51 words

They had been running two systems side by side for eleven months. One held

the orders, the other held what was actually on the shelf.

The switch took a weekend. What this video covers is the first Monday, and

the number the floor team quoted back at the end of it.

approving commits 2:07 to this month

Review is a switch on the template. Turn it off and this screen never appears: the script goes straight to the voice.

After you approve, you can stop watching.

  1. probe00:04done, artifact kept
  2. scene detection00:38done, artifact kept
  3. script00:21done, artifact kept
  4. script review3d 04:12waited for you
  5. voiceover00:33done, artifact kept
  6. captions00:11done, artifact kept
  7. render00:47done, artifact kept
  8. seo metadata00:06writing metadata

There is no percentage bar anywhere in obclip. The job is in one of eight named stages and the interface says which. Close the tab and the state is still true when you come back.

notificationtoday 09:14

Warehouse module release note is finished

to you@yourcompany.com

Still frame from the finished video
runtime
1:52
aspect
16:9
template
release note
minutes charged
1.9
support id
obc-7f42-19c

The mail carries the finished file.

Plans built around finished video minutes, not seats.

18 of 30 finished video minutes used

resets 1st of the month, 00:00 UTC

Pick the amount of finished output you need each month. Every plan is built around the same obclip promise: raw footage in, final video out.

essentials

€6.50/ month

5 video minutes / month

For solo operators who need a small, reliable video production allowance.

  • 5 finished video minutes per month
  • 2 min max upload length
  • 1 GB storage
  • 1 workspace
See the full plan

starter

€16/ month

30 video minutes / month

For teams getting started with automated captions, branded narration, and clean exports.

  • 30 finished video minutes per month
  • 2 min max upload length
  • 10 GB storage
  • 2 workspaces
See the full plan

custom

Custom

Higher volume, team, or agency needs

For teams with higher volume, custom storage, or workflow support needs.

  • Custom video minute allowance
  • Custom storage
  • More workspaces
  • Migration or workflow support
  • Direct onboarding
Talk to us

Common questions about the obclip pipeline.

What is obclip?
obclip is an automated video production pipeline for teams that need to turn raw footage into finished videos on a schedule.
Who is obclip for?
obclip is built for developers, operators, agencies, and content teams that want one staged workflow instead of separate tools for scripting, voiceover, captions, rendering, and metadata.
What comes out of the pipeline?
A typical run produces a scripted, voiced, captioned, rendered, and metadata-packaged video that is ready for review or publishing.
Do I have to review every script?
No. Script review is a switch on the template. Leave it off and the job runs from footage to finished file without stopping for you. Turn it on for the videos where the wording matters, and the pipeline holds after the script until you approve or edit it.
Why does the staged workflow matter?
The staged workflow makes runs inspectable and repeatable. Teams can review artifacts, rerun downstream stages, and keep production moving without rebuilding the whole job from scratch.

What obclip does not do

It does not clip.

No shorts, no highlight extraction. One finished video per job.

It does not publish for you yet.

You get a file and its metadata. Where it goes is your call.

It does not generate presenters.

No avatars, no synthetic faces. You shot the footage, it stays yours.

It is not a timeline editor.

There is no cut-level control. If you want to art-direct, use an editor.

Commit to a cadence you can keep.