Short video ready for distribution

How do you turn scattered raw footage into a short, understandable and verified video with AI assistance?

1250 € per session, excl. VAT

In short

In a half-day session of 3 hours you assemble a short video of 60 to 90 seconds from fictional raw footage. AI helps suggest an editing plan and refine the subtitles; you retain editorial control, check the rights and assess the exported file before distribution.

Why this training?

A short video quickly loses its usefulness when the sequence of shots obscures the message, cuts alter the meaning or subtitles become unreadable. Editing should prioritise understanding rather than an accumulation of effects. AI can suggest a structure based on a transcript, but its suggestions must be checked against the footage. This course combines editorial assistance, hands-on editing and checks on the result, without delegating responsibility for publication. Running case study: The fictional association « Avelune Réemploi » wants to explain how to drop off a small appliance at its workshop. Using six raw clips, a transcript and fictional instructions, participants create a short video for its reception screen that can be understood without sound. Application at work: For a forthcoming authorised video, reuse the brief, AI instructions and checklist. Compare the first version with the corrected version: duration, subtitling errors, understanding of the steps and time spent on revisions.

Objectives of this training

  • Define a single message and a distribution format.
  • Select useful raw footage and develop an editing plan with AI.
  • Assemble shots and adjust cuts, pacing and audio levels.
  • Correct subtitles while preserving the meaning of the speech.
  • Export a short video and measure its quality using an assessment grid with observable criteria.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communications officers producing short videos for internal or public channels.
  • Staff who occasionally create short explanatory videos from existing footage.

Prior skills

  • Have previous experience of making simple requests to a generative AI assistant.
  • Be able to import a file and make a simple cut in video editing software.

What you will work on

  1. Define the scope and ensure safe use of media

    Objective: Establish the message and select usable resources. Audience, target screen, aspect ratio; image and music rights, individuals’ consent, confidentiality and conditions for importing into AI.

  2. Prepare the assisted editing plan

    Objective: Turn the transcript into a verifiable editing proposal. AI prompt specifying the message, duration, available shots and a prohibition on inventing content; locating passages using timecodes.

  3. Assemble and make accessible

    Objective: Produce a coherent first version. Shot placement, cuts, continuity, pacing, audio levels and subtitles. AI assistance according to the functions actually available; a manual alternative is provided.

  4. Check and export

    Objective: Correct defects before any distribution. Export settings suited to the channel; full viewing; notices identifying synthetic content according to the context and applicable rules; risks of deceptive deepfakes.

What you take away

  • Short video brief and timed editing plan.
  • Editable video project with organised media.
  • Exported short video of 60 to 90 seconds with subtitles.
  • Completed quality assessment grid and record of rights, consents and notices.

Method and assessment

  • Short instructional inputs followed by immediate practice using the running case study.
  • Individual work followed by work in pairs, with peer review of the videos produced.
  • Guided group review: what works, what needs correcting and what remains a human decision.
  • Preparation: Have a workstation with headphones, video editing software and a tool authorised by your organisation; check importing, subtitling and exporting before the session. Download the fictional or anonymised training materials: raw footage, transcript, brief and rights record; no unauthorised personal media are used.

Observable criteria: The short video meets the target duration and presents the steps in the correct order. ; The cuts do not distort the instructions and no shots remain missing. ; The subtitles are synchronised, readable and free of errors that alter the meaning. ; The export plays in full, with intelligible speech and no audible clipping. ; Each media item has a documented rights status; the necessary consents and notices are identified..

Frequently asked questions

Does AI automatically perform the entire edit?

No. It assists with certain decisions; assembly and approvals remain under the participant’s control.

Can internal raw footage be used?

Only if they are anonymised and authorised, with permission to import them into the chosen environment; otherwise, use the fictional materials.

Do we create an avatar or an imitated voice?

No. No deepfake is produced; consent, the risk of impersonation and the labelling of synthetic content are examined.

How is the resulting quality assessed?

By viewing the export, checking the subtitles and testing understanding, with a comparison of discrepancies before and after correction.

Last updated: