Points for speakers' preparation and recording set-up:

On video tutorials made at CERN

  • Create a lecture in the e-learning indico category for every rehearsal to keep trace of the process and notes for future use. The lecture classification of the indico event is important if you wish to request recording.
  • Make sure the indico Description field contains all the necessary and sufficient information a candidate video viewer needs to know to be interested and understand your video. This information will be used for the CDS Abstract content. Once your video will be published in CDS, you'll need to open a SNOW ticket to cds support every time you need to change it.
  • Book recording in indico, if the video is done by the CERN audiovisual support. Ask, in your recording request, that the support experts test the video performance on mobile devices as well as computers for video speed, sound, image and slide sync quality before publishing on CDS. To book a room which has recording facilities via the rooms search in Indico - https://indico.cern.ch/rooms/rooms/search/ by ticking 'webcast/recording' and unticking the special attributes.
  • Display the tailor-made e-learning logo before starting and the CERN logo at tutorial closing. Scroll down to this page's attachment for ready cover slides. NB! If you record your tutorial by yourself, e.g. with ActivePresenter or QuickTime, make sure the entry slide with the e-learning logo lasts for, at least 5 seconds. This is indispensable for the CERN audiovisual team, in order to position the first frame in the CDS collection index. The cover/corporate slide was equipped with music following this CERN IT CDA interval voting.
  1. First slide with music
  2. Last 3 slides

  • Browser and terminal windows should be in one of the CERN official languages. Starting point should be a CERN web page.
  • Use a screen capture tool (depending on your OS: ffmpeg, quicktime, ActivePresenter, shotcut kazam,...) for your rehearsals. By recording yourself, you can see what needs improvement. See here the documentation on such tools. Some demos can be found in the VideoLibrary.
  • To publish a self-made video in CDS, it is best to record in mp4 container with H.264/MPEG4 codec. Recommended full settings for self-made tutorials are: 1080p@5Mbits H264 codec audio in AAC 96 kbps .
  • For self-made videos please use the a standard resolution and aspect ratio. The CERN IT transcoding profiles are for aspect ratio 16:9 or 4:3 equivallent to 1920x1080
  • It is recommended that the final version of your video to be published in CDS.
    • For each CDS video record the number of viewers appears on the page.
    • Statistics can be obtained by the CDS experts via SNOW requests by the content owners.
    • Comments are possible via the Discussion tab available over the frame of every CDS video (Example).
    • You may embed the video in your documentation. See an example from the Indico documentation. See also SNOW KB article on how to do this. Basically, just enter in your document source the statement:
<iframe width="576" height="360" frameborder="0" src="https://cds.cern.ch/video/XXXXXX?showTitle=true" allowfullscreen></iframe> 
<div style="float:right;position:relative;margin-top:-250px;">
<video controls="" style="width:350px;" width="350"><source src="https://mediastream.cern.ch/MediaArchive/Video/Public/Conferences/201x/XXXXXX/XXXXXX-2000-kbps-1280x720-23.98-fps-audio-96-kbps-44-kHz-stereo.mp4" type="video/mp4" /></video>
where XXXXXX is the number of your video in CDS.
NB! If your video is recorded by the audiovisual support and contains a camera channel, i.e. face+slides/screen+voice, i.e. not only slides/screen with voice over, you may need to ask, in your SNOW ticket for video publishing, the isolation of slides+voice channel, used when embedding the video in a web page, to avoid any delay between channels, outside CDS. Example: https://twiki.cern.ch/Edutech/CMSVirtualVisitsInstructions contains voice and slides (no face), while the relevant CDS record contains the speaker.

General advice for any short video recording

These points are available as slides and a 3' 34'' online tutorial in CERN CDS and linked from indico.

  • Write down every word you plan to say. Attach it to the relevant indico event. Example the lhcathome tutorial's text by Karolina Bozek.
  • Give a structure to your text and say where you are as you go on, e.g. "This tutorial is about this cool feature... I'll explain what it does, how to install, how to configure, how to start, stop, monitor..."
  • Use a teleprompter (like a TV news' broadcaster). !! At least until the 50th rehearsal it is needed. This free-of-charge teleprompter on the web allows very easy upload of your plaintext file.
    See here why
  • Use a microphone. The benefit is not negligeable!
  • Rehearse, rehearse, rehearse. Use colleagues familiar with the domain as your public to obtain feedback.
  • Measure the time. Short tutorials should not exceed 5' to maintain audience attention. Dropping from 8' to 4' requires many rehearsals.
  • Check what is already available on the web, as examples to learn from.
  • Make sure everything you type in the video respects security rules. No personal emails, logins, passwords etc. The CERN IT recording team cannot edit video to blur text.
  • Avoid typing commands 'live'. Rather make slides, screenshots or web page tabs with the commands on.
  • Redact any screenshots by blurring out any details which are not absolutely necessary for the topic addressed. Be extra careful with logins, filepaths and any other hackable info.
  • If you have slides, select the 16*9 landscape format for the slide dimension and point on your slide.
  • Make sure the web pages of the application you present are clear for new-comers. Else, talk to their webmasters before recording the tutorial.
  • Remove desktop icons, browser tabs, other applications (email, skype) that can pop-up images and sounds while you record.
  • If you present a tool that works on multiple Operating systems, make a generic introduction and shorter episodes per Operating System.
  • The camera should show the speaker first with a corporate picture in the background and then the slides, web pages and command lines of the tutorial.
  • Make sure no items get in the video belonging to commercial brands (e.g. water or soda bottles). Laptops are fine.
  • Write "Do NOT disturb" on the door when recording to avoid people entering during your best try.
  • Use a normal chair (no wheels, not twisting) to minimise body movements during the video. Keep hands visible.
  • Explain what happens if the package you are describing is already installed
  • When mentioning a web site make sure it is on the screen at the same time
  • When encouraging users to contact you, show them where is the link for them to write
  • Say: *READ* the terms of use before agreeing, when clicking on licence agreements during the tutorial
  • SMILE while speaking
  • When skipping installation details say if all is standard installation process for the given platform or make a screen capture video (ffmpeg or quicktime or ...) and fast forward. Make sure this secondary video is on the same web page as the main one (indico, CDS and/or Video Library pages for CERN tutorials).
  • Some browsers need extra plugins to show certain filetypes. This is mainly due to copyright issues. The way around this is to include 2 or 3 versions of the same video in the web page dedicated to a given tutorial.
  • Use a screen capture tool for your rehearsals. By recording yourself, you can see what needs improvement.

More advice

-- MariaDimou - 2016-11-17

Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf Ai-Media-Captioning-DIY-Style-Guide.pdf r1 manage 1479.9 K 2018-06-20 - 10:49 MariaDimou How to create professional captions by company Ai
Edit | Attach | Watch | Print version | History: r10 < r9 < r8 < r7 < r6 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r10 - 2019-04-01 - MariaDimou
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    ELearning All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback