If you’re sure you haven’t missed anything, try extending the lead-out a little. ![]() If it’s still too short, make sure you haven’t missed any speech that should be part of the line. This one should be longer, ~400 or 500 ms should be okay. This is called lead-in, and it gives the brain time to catch up to the fact that a new line has appeared. Now we move on to fine timing, where we make the subs comfortable to read.įirst, add a little buffer at the start of every line. When you’re all done, the script should look something like this. If it’s only a little bit of overlap at the start and end, just leave it. If they’re both equally important, add an overlap style (your normal dialogue style, but a bit different - usually a bit of (dark) colour on the border) to whichever one started speaking last. If one of the speakers is less important, like an announcement in a train station (versus an actual conversation), add to the start of their lines to move them to the top of the screen. (don’t buy audiophile headphones just for timing) This will probably much easier with nice audio equipment, but you can get by fine without. Then you “just” have to pick out where the different voices start, and time like normal, ignoring the other speaker. Two people speaking at the same time in the sub grid The easiest way to deal with this is by putting each speaker’s lines into their own “block”, where all of their lines come one after the other in the grid, regardless of the order of the whole conversation. Probably most likely is two people talking at the same time. There are a few edge cases you might run into. Rinse, move on to the next line, and repeat. Don’t worry about being millisecond-accurate (it’s only rough, after all), but do try and get it close. Check you got it right by playing the entire line, and/or by using the 500ms before/after buttons ( ) to make sure there’s not anything you might have missed. Left click for the start, right click for the end, or just click and drag. Go to the first line of dialogue, and set the start and end times to exactly where (you think) the line starts and ends. Okay, now we’re ready to actually start timing something. I have no idea why anyone would ever have it off. You can change it in Options > Interface > Colours, under Audio Colour Schemes.)Īlso, turn on auto-commit. I’m partial to Green, myself, and that’s what you’ll see in this guide, but it’s really just personal preference. (There are two options for what colour it can be: Icy Blue, and Green. With practice, you can pick out actual voices from most other background sounds, which you just can’t do on a waveform (because it only represents volume). It may look confusing, and might take a little while to get used to, but I promise you it’s worth it. Higher up means higher pitch, brighter means louder. This is called a spectrogram, and it’s a way of representing sound in an image. If what you’re seeing now looks anything like the above, you should change it. All we care about for now is the audio, the keyframes, and the script. Now, the first thing you’ll want to do is get rid of that video you just loaded. The lines under the video that mark out keyframes The change will be much less obvious, but, assuming all went well, you should see the little lines beneath the video change. Load your keyframes in much the same way (but with Open keyframes, obviously). Load your video in with Video > Open Video., or by dragging it onto the subtitle grid. I assume you can figure out how to open your script. TODO: finish this paragraphīoth of these methods should produce a file ending in _keyframes.log. ![]() You’ll need a copy of FFMPEG (almost certainly in your package manager on linux), and scxvid-standalone, which you’ll have to compile. Drag and drop your video onto it, and wait.Īnywhere else, you’ll have to do a bit of setup first. If you’re on windows, use this batch script, which has everything you need bundled in. You’ll almost certainly have to generate your own at some point, so get learning how to do it out of the way now. Your video will have keyframes already in it, and they might be decent enough, but they might also be completely useless. These are very important if you want your timing to look good, which I assume you do. If you spot something wrong with any part of this guide, feel free to yell at me on discord. You’ll also need Aegisub, if you don’t already have it. ![]() The easiest way is to grab some random release off the cat site, extract the subs, and set all the timestamps to 00:00.00. ![]() Timing: Lining up the words of the script with the words of the audio, preferably in a way that looks good and is easy to read.īefore you do anything else, you’ll need to get a video, and a script. You shouldn’t read this yet lol, it’s still very WIP
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |