VarsityEditor
Well-known member
- Messages
- 117
- Reaction score
- 143
- Trophy Points
- 48
Overview:
Terminology:
My personal view:
Here’s an overview of the various areas of relevance. I’ve added a few notes, but this is mostly just meant as a starting point.
I have my personal experience with adding subs to edits, but of course it’s limited to the platform/software which I use. This is something which will vary for others. I’d be grateful for input from anyone who has experience in doing this.
A— USE CASES:
B— SOURCES OF THE CAPTIONS:
C— NLE SOFTWARE:
D— DEDICATED CAPTIONING SOFTWARE:
E— SHARING CAPTIONS:
Any input that anyone can give on any of these areas would be gratefully received! I will add more as time and experience allows...
I think a good long term goal is to steer towards increasing the prevalence of edits having subtitles. Optional subtitles/captions are near-universal on commercial releases, but seem to be mostly ignored for fanedits, treated as a luxury item which usually isn’t worth bothering about.
The purpose of this thread is to try to gather experiences, tips, ideas, advice, dos & don’ts, pitfalls, and best practice together, with the general aim of eventually being able to produce a “how-to” guide.
I have no expectations that five years from now, 100% of edits will have captions included, or that FE.org will mandate inclusion of accurate subs as part of the requirements of a completed edit — I just think that it’s a good direction to move towards, and that the first step is to start to centralise knowledge to establish and spread best practice. Maybe in a year or two there will be a pretty straightforward guide available that new (or old) editors can be pointed towards to streamline the process of including subtitles. Maybe in five years, 25% of new edits on the site will have captions rather than 10% (I’m just making up numbers but you get what I mean).
Terminology:
As I understand it, captions typically contain more than just the words that characters are speaking – they also include descriptions of audio ("door slams", "dog howls" etc) which help the viewer get a complete picture of what is happening if they can’t hear, while subtitles generally just give the words spoken, the main purpose being to provide a translation for the viewer, but otherwise assuming that the viewer can hear music and SFX.
This thread is just meant for general hashing out of ideas about the editing process, and I’m using “captions/subtitles/subs” interchangeably.
I’m also talking about non-hardcoded subs here. As in, the primary viewing experience is without subs, but the viewer has the option to switch on “English Subtitles”.
My personal view:
When I started editing I had no intention whatsoever of including subs. Fanedits are already a niche interest, and adding subs seemed like a completely unnecessary niche of a niche. Editing is enough of a detail oriented, time-consuming task as it is, and the cost-benefit analysis for adding subs just didn’t seem to make sense.
Now that I’ve done a lot of editing and am in the process of adding subs to everything, I wish that I had just included them from the start! It would have been so quick and easy, and is so handy while editing.
While we typically think of subs as something to benefit the viewer, but of little benefit to the editor, I have found it editing with captions in place to be a massive bonus. It’s ten times easier to scrub through the timeline looking for a particular moment in a scene when the words that they are speaking are appearing on the viewer. While I’ve found it to be a massive positive difference for the editing process, the main motivation for doing this is to make the final edit a more professional, complete product, which gives the viewer the best experience, just as we care about having smooth scene transitions or even audio levels.
Here’s an overview of the various areas of relevance. I’ve added a few notes, but this is mostly just meant as a starting point.
I have my personal experience with adding subs to edits, but of course it’s limited to the platform/software which I use. This is something which will vary for others. I’d be grateful for input from anyone who has experience in doing this.
A— USE CASES:
A1. Including subs from the beginning of a new editing project.
For me, this is the ideal situation. If the standard captions are added to the unedited movie(s) before you start chopping it up, then you can basically just do all editing as normal from that point onward with effectively zero added work, and have perfect captions at the end.
A2. Adding subs to a current/in progress editing project
This is what I’m currently doing a lot of. It has consisted of downloading available SRTs online to fit the source video, importing them to my NLE project, and moving them around as necessary to fit the edit. Nowhere near as tedious or time consuming as I expected, a one-time operation of roughly 30-60 mins for a 2+hour project, depending on complexity.
A3. Adding subs to a complete edit/video
This might be done if somebody wants to add subs to an existing video – perhaps something they no longer have the project files for, or even something produced by someone else.
B— SOURCES OF THE CAPTIONS:
B1. Sourced from the physical media rip
I wish I just did this when I made my video rips in the first place, but no bother, just use option 2–
B2. Sourcing external SRT text files from websites
This is what I’m doing, and if available, it’s much better than the following options–
B3. Manually
The old fashioned way. Ears listening and fingers typing. Insanely time-consuming, and an option in name alone!
B4. Using AI speech recognition tools to automatically create captions
I started off trying to use this approach. In my experience, while it takes 99% of the work out of option 3, the results are nowhere near usable. Many captions will be mistimed, and many have mistakes (wrong words, spelling etc) which need to be manually corrected. Even when only 5% need correcting, it's still very time consuming compared to copy-pasting available SRT files.
C— NLE SOFTWARE:
C1. Final Cut Pro X (Mac)
This is what I use. Adding hundreds of captions to the timeline will slow things down and have a noticeable performance hit, but that’s only when the captions are shown in the timeline. When switching off their visibility (in the timeline, not on the screen), there is no loss in performance.
C2. DaVinci Resolve (free/pro)
I don’t normally use this, but dabbled around testing the AI captioning capabilities (pro version only). They are pretty good, but still require lots of manual correction.
C3. Sony Vegas
I don’t know anything about this program, I just include it here as I’ve noticed many people mentioning it on these forums.
4. Adobe Premiere/Others…
D— DEDICATED CAPTIONING SOFTWARE:
D1. Subtitle Edit
I haven’t used this, but have seen others here talking about it, apparently it does a good job when you need to manually edit the subs.
D2. Others…
There are many tools/plugins/apps around now — mainly marketed toward the TikTok/Instagram influencer sphere — to create real-time captions from an audio/video file. I’ve tried a couple, and found them less good than the DaVinci Resolve version. Maybe good for a 60 second TikTok, but not fit for purpose for an hours long movie (see point B4).
E— SHARING CAPTIONS:
E1. Hardcoding/burned-in to the video
Obviously not recommended as a general approach (except for non-English/non-primary language dialogue)
E2. Packaged with video in container format (eg MKV instead of MP4)
The subs are contained within the single video file, and can be switched on in the user's video player.
E3. Separate file
Eg .SRT text file (other formats available). A small text file which can be shared along with the video. Alternatively, they can be easily uploaded and hosted on the many websites for hosting/cataloguing subtitle files (same sites as used for source in B2), meaning they are accessible to viewers.
Any input that anyone can give on any of these areas would be gratefully received! I will add more as time and experience allows...