Hi . Im on to prepare a big Documentary Project with lots of interviews in mixed Mac/ PC environment and i would like to ask the community for input and experiences
1. Challange. Its going to be edtited with FCPX ( of course but a lot of Offloading, Media, Tagging, etc will be done on PC'S- we are going to get a Areca Raid5 for the material, But What format ? HFS+ and a Software for the PC or X-fat for both or NTFS and Software like Paragon for the Mac with FCPX . any experiences and suggestions ?
2.) for keywords and tagging a looked at KYNO which i think would do a good job, since its cross Plattform and the Metadata goes to FCPX . I am also going to transcript the Interview Material, Probably manual , since the Dialects in Austria my be hard for Software . Is there a possibility to combine this with KYNO so that i can find sentences in FCP later when editing. i saw some clips here in fop.cp ether with Notes, or in Titles . is there a good proven Workflow with KYNO ? or other software to accomplish this ?
How many total interviews, how many with external audio, and how many are multi-camera?
The strongest part of FCPX is organizing and tagging the material. For multicam you normally sync those then keyword/rate the MC clip, not the parent clips. Keywords and ratings done on the parent clips are not passed down to the MC clip.
Re disk format, I'd be hesitant to use anything besides HFS+ for locally-attached storage with FCPX. I use Paragon NTFS for data interchange also exFAT but I would never trust those for a large FCPX library. I'd be afraid some dependency in FCPX for hard links, extended attributes or something else wouldn't work right.
I haven't used KYNO recently so I don't remember how well it supports range-based tagging. However if you have much multicam material it doesn't really matter since the MC clips must be built within FCPX and those don't inherit ratings/keywords from the parent clips.
It would almost be better to create proxies on a powerful Mac then get an inexpensive MacBook Air, use a proxy-only workflow and have your assistants use FCPX to keyword and rate the material on that. Those can then be shipped back to the larger Mac via a lean library or XML. I have trained assistants with PC-only experience to do this in a few days.
Maybe someone with recent KYNO experience in a cross-platform FCPX situation could better comment.
Thanks. for the raid we will go for an Areca 8 bay on Thunderbolt3 with hfs+ format and paragon on the PC and a LTO 8
As for Kyno. is there a way to tag / combine transcripts ? Or a good way to combine Kyno with the lumberjack system. Kyno is cross platform, but how can the transcript be combined with the video? range based in the notes field ?
I saw a very interesting clip on the luberjack builder . transcript based pre editing,.. what i saw its mac based.
anyone have experience with that ? Is Kyno needed at all with the lubernjack system ? or are both needed/ or have an advantage
thanks for any thoughts and experiences.
The administrator has disabled public write access.
Big Project preparation questions, Mixed Environment Mac / pc , Kyno & Transcribe
23 Mar 2019 14:54 #99408
...As for Kyno. is there a way to tag / combine transcripts ? Or a good way to combine Kyno with the lumberjack system. Kyno is cross platform, but how can the transcript be combined with the video? range based in the notes field ?...I saw a very interesting clip on the luberjack builder . transcript based pre editing,.. what i saw its mac based....anyone have experience with that ? Is Kyno needed at all with the lubernjack system ? ...
There are various ways to handle interview transcripts, whether for foreign or native languages. The simplest way is manual -- use timecode per line or per range of transcript text. Whoever transcribes it can enter the timecode. NCH makes pedal-controlled transcription software. I don't know if it inserts timecode per line, per range or what: www.nch.com.au/software/dictation.html
Besides Lumberjack's Builder, there are other voice recognition transcription systems, but (as you said) I don't know how sensitive they are to dialect. Some of them produce XML which can integrate with FCPX. I have not tested those. I think they all rely on 3rd-party transcription services which might cost $0.50 to $1.00 per minute. Besides Builder there is Transcriptive from Digital Anarchy: store.digitalanarchy.com/videoplugins/263-transcriptive.html and Scribeomatic from CoreMelt: coremelt.com/products/scribeomatic
For foreign language transcription one method is have the translator prepare the transcript, then verbally record the translation while listening to the foreign language. With practice this can be done while maintaining +/- 1 to 2 sec sync. That audio file is then added as a separate lane to the original interview, sync checked, then the editor can make editing decisions based on their language and fade in/out of the translated language. Typically the final edit would require re-recording by an age/gender-specific voice actor.
To reduce work, some documentary producers simply allow the translator to pick out selected statements and transcribe only those. Of course the translator might not have the artistic vision of the overall project so might not pick out the right statements.
Re KYNO, in general I think the best approach is have the assistants organizing the material use FCPX. It is hyper-fast and can be imported using "leave files in place". The keywords, ratings and sync'd multicam clips can be exported to FCPX via XML. The work of multiple assistants can by merged with MergeX: www.merge.software. If the material is 4k H264, proxies would likely be required for good scrubbing performance. That is another complication. In theory the assistants could be supplied with proxy-only media and use a lower-end Mac but proxy-only workflow is yet another complication. The details are documented in Ripple Training's FCPX 10.4 Media Management tutorial: www.rippletraining.com/products/final-cu...-final-cut-pro-10-4/
However that is not always possible. KYNO is a great product. In theory you can mark clips and keyword ranges (aka sub-clips) using KYNO on a PC and export those to FCPX on a Mac. KYNO has a very fast playback engine and can "flatten" folder trees giving rapid access to the media. This workflow should be robustly tested, you shouldn't just jump into it. There is a recent FCPX Radio podcast with updated info on KYNO. It is very informative and anyone interested in KYNO should listen: fcpradio.com/episode082.html
Another complication regards multi-camera material. This can include single-camera with external audio, since with FCPX it can be best to use a multi-cam clip for that. Sam Mestman mentioned the advantages of using multicam over sync clips here:
When rating or keywording multi-cam clips it's best to do this only on the MC clip itself, not the parent clips. Ratings and keywords are not inherited by the MC clip. In our workflow we always reject the parent clips after making the MC so nobody will accidentally keyword/rate those or edit them into a timeline. So if you invest effort on tagging material (either within FCPX or KYNO), then create the MC clip, those tags will not be usable.
Regardless of how you rate/keyword material, it's important to have a consistent approach. You don't want one assistant using "happy" and another using "laughing". You don't want more granularity than is necessary, else the final editor will get confused and not use it. It is easy to spends lots of effort on a grandiose tagging system which never gets used.
One of the hardest things for a documentary editor is finding usable b-roll to cover interview statements. If your assistants could use keywords on clip ranges which correspond to the interview statement needing b-roll, that could help. Likewise the reverse -- if you have a compelling b-roll shot, how to find if any interview statement mentions that. With trained assistants you can theoretically have them do much of this work for the final editor.
If you use multiple libraries there is a significant "gotcha" about dragging clips between libraries. The best practice is don't drag or copy bare clips or projects across libraries but do it inside an event. Sam Mestman discusses this from 06:30 to 11:00 in the below video, but you may want to watch Sam's whole talk from 02:00 to 11:00. He demonstrates this on a Lumaforge NAS but it's not unique to a NAS.