DAM video thoughts

A couple of recent conversations have prompted me to think a little more about digital asset management for video.  I have been doing some DAM work already, but it has been with still images and I haven’t yet had to deal with the added complexity of video content.  The obvious issue with video is the size of the files, along with the fact that it (usually) contains audio, including speech, and thus requires so much more storage space and accompanying metadata than still images.

The first conversation was with a former colleague of my wife’s who is now handling sales of a suite of products that supports litigation teams.  One of the services they offer is video deposition support.  This includes taping the deposition and supplying a transcript of the deponent’s testimony with time codes inserted so that the user can easily locate the video footage of the testimony in question.  Depositions are so important in litigation, although you don’t usually see them portrayed in films or on television, because they consist of sworn testimony equivalent to testifying in open court.  Therefore, attorneys need to analyze deposition testimony in preparation for trial.  You can imagine how useful it would be to have the video along with the searchable text of the deposition.

The second conversation was with a married couple at a social event my wife and I attended last night.  The woman is a master sommelier who has a website where she features educational videos about wine.  Her husband handles the videography for this content.  I mentioned that I am involved in digital asset management and explained a little more about the issues DAM tries to resolve and they immediately got it.  They gave the following example of an instance where they could use some help classifying the footage they shoot: Say they are producing a clip about a certain type of wine, but during the video shoot someone brings up a topic that was not previously planned.  This couple described a scenario in which they would like to be able to return to that footage at some point in the future, possibly because they are now producing content about the second subject.  But if they are classifying their footage by the primary topic of the shoot, it is extremely difficult to go back and locate the clip they want to retrieve, because at that point it becomes a question of remembering something that took place weeks, or even months, prior.

So I started thinking about how to apply the concept of a time coded transcript to the wine-related content.  I know this is probably nothing new to people who deal with a lot of video footage, particularly people who work in scripted television or film production where the content is predetermined.  But in situations where the content is more free flowing, like in interviews or casual conversations, you are not working from a script and so you don’t know beforehand all the subjects that may come up.  This same scenario is at work in a deposition (although most attorneys certainly try to guide the conversations as much as possible), or in a conversation about wine.  People will say things without planning them in advance, based on where the conversation leads.

The question then becomes: Is it feasible to have someone transcribe the content for the wine productions and then feed it into a DAM system where the text is searchable?  I like to think that it is.  Perhaps we need to find a court reporter who has a little bit of extra time and would take wine in trade for their services.  After the transcription is complete, of course.

Advertisement