Why logging video clips isn’t always the most efficient way to cut down edit time.
In the old days, when Betacam was king, shoot ratios were restricted by how many tapes you could afford to take on the shoot. Now, thanks to digital workflows, production teams have an almost limitless capacity to record footage and shoot ratios are skyrocketing as a result. But the cost of recording more footage isn’t limited to the shoot – more footage means you need more time to transfer the media, more storage space and more time in the edit suite to cut it down. Logging your video footage is generally considered the best way to whittle down material before post, but is it the most efficient way to prepare dialogue-based material for edit?
Logging video clips to reduce costs
There are two options for logging video clips that can help reduce some of the costs that result from high shoot ratios. You can either log-as-you-go on the shoot or review and log your footage once the shoot is complete.
Shoot logging is especially useful if you’re going to be recording multiple takes of one shot – like if you’re working with talent that need to deliver specific lines or if you’re filming choreographed activities that need to be perfect. Logging the in and out timecodes of the good takes on the shoot means that you can dispose of the outtakes before ingesting the footage for edit.
An alternative is to log your footage after the shoot. This approach means that you’ll have to first import all your shoot footage (so you won’t necessarily save ingest time or storage space) and allocate time in between the shoot and edit to go through all the media, but it’s a good way to familiarise yourself with your rushes and to start selecting shots for the edit. Depending on what system you’re using, you should also be able to organise your media into bins and tag clips with metadata to make it easier to search for specific shots in the edit.
Both these approaches to logging can be used to find clips where specific actions take place, shots that feature particular items or to identify content themes, but they aren’t ideally suited to building narratives from interview material. That’s because logs don’t typically include full transcriptions of the dialogue captured in video interviews – which means they can’t be used to search for specific phrases. For example, you may be able to search for the clip where the interviewee is asked about the price of eggs and find the shot of a box of eggs, but you won’t be able to find all the instances where anyone mentions the word “eggs.”
Using artificial intelligence to find soundbites
Recent developments in automatic speech recognition have provided some quick and cost-effective solutions to searching for key words in video interview footage.
Some NLE products have integrated ASR technology into their software, for example, Avid’s Phrasefind and Dialogue Search products are built on Emmy award winning Nexidia technology to provide editors with the ability to search for key words spoken across projects. But relying on these tools to find key interview clips means that all your footage would have to be ingested and stored until you get into the edit and you’d spend valuable edit time searching for material and planning storylines instead of doing this prep beforehand. And, because some of these tools are based on phonetic search, they rely on your ability to “think like a microphone, not like a dictionary” to get accurate results.
ASR is also being used to bring down the cost of transcribing rushes. Transcribing your source’s audio track provides a text-based version of your interview or dialogue, making it easy to search for key words and phrases before the edit and plan your storylines. But, while ASR technology is making transcriptions more cost-effective, this automated software is not yet able to provide the same level of accuracy as human transcribers . And, if the key words and phrases you’re searching for have been inaccurately transcribed, then they won’t show up in your search.
Using accurate transcriptions to create searchable data
When it comes to dialogue-based productions, creating accurate, digital transcriptions of your source material is the first step to efficient editing. The second step depends on your preferred post-production process, you might choose to:
• Use simple software like Word to search for key words and phrases in your transcription and cut and paste segments to create a paper-edit of your storyline. High quality transcriptions mean you can trust that your search will be accurate.
• Import your transcription into a logging and rough-cut tool so that you can combine the power of logging and transcription metadata to speed up your searches and build a stringout. Take 1 integrates with the Blackbird cloud editing platform to provide a jump-start on the edit in this way.
• Import your transcription into Avid and use ScriptSync to automatically link the text to your video clips. Take 1 can deliver transcriptions in ScriptSync compatible files so that you can easily build timelines based on key words in your transcript and search with absolute confidence.
Get in touch with Take 1 to find out about our pre-edit transcription packages.