how do i efficiently attach audio to an image

haroldstork, 1 year ago

This might depend on where you’re uploading/how you’re playing this file, but you could add a thumbnail to the audio file? I know that vlc and mpv will play your audio file and show the thumbnail, but I’m not sure if YouTube would take that. Not sure if this is exactly what you’re looking for but it is pretty efficient.

hinterlufer, 1 year ago

In kdenlive, the following settings work well for me (you can transfer the options to ffmpeg cli as well if you prefer that):

f=matroska movflags=+faststart vcodec=libx264 tune=stillimage progressive=1 g=1000 bf=2 crf=%quality acodec=flac ar=48000

For reference, I get a 3.7 GB video with a duration of over 5 h @4k resolution. The audio itself is already 3.7 GB and it’s just a still image. For CRF, set something around 23, that should do.

bbbhltz, 1 year ago

Recommended YT settings might be helpful here.

What command have you been using?

andrew_bidlaw, 1 year ago (edited 11 months ago)

Can you intentionally set various bitrate (VBR) with a big difference, like from 16mbps max to 1mbps min? Constant bitrate can be your problem. Premiere had this on media export menu, I can’t remember where others have this. And, if you want to upload it to youtube, see what size it has on a private upload by downloading it - they reencode every video themselves to their uniform standard, so maybe you don’t even need to bother with that.

turkalino, 1 year ago

Are you sure your rendering settings are correct? It sounds like the video isn’t being encoded at all. Video encoding works by storing a frame in full quality every couple seconds or so. For the rest of the frames, only their differences from the previous full-quality frame are stored. But from what you describe, it sounds like the latter sentence isn’t happening

jackpot, 1 year ago

oh i didnt think this is how it worked, i ended up using ffmpeg to make a one frame a second video but it still looks blurry??

nekusoul, 1 year ago (edited 11 months ago)

Since the end goal is to post a video to YouTube, you will have to create a video file. Personally I would probably just be lazy and upload the large file, since YouTube is going to reconvert the video anyway.

That said, to optimize the file you need to know how videos work, specifically key frames. Speaking generally, when a video gets encoded, it doesn’t add the whole image for each frame. Instead, it only does that when the current frame is a key frame, and then only stores the difference to the previous frame for every regular frame. There’s a lot of different strategies when placing keyframes, like every X seconds, when the scene changes, or both. This is usually you can change somewhere in the encoding settings of the application you’re using. You will need to use a codec/format that supports interframe compression though, so avoid AVI and MJPEG.

So the TL;DR is: Try to decrease the amount of key frames as much as possible, maybe even down to only one if possible.

gray, 1 year ago

superuser.com/…/combine-one-image-one-audio-file-…

luthis, 1 year ago

The answer is always ffmpeg

jackpot, 1 year ago

im confused sorry, i tried the libx265 codec but the image still has blurry edges??

jws_shadotak, 1 year ago

Are you able to add it through the file properties menu?

jackpot, 1 year ago (edited 1 year ago)

probably? would that be able to be sent to people / uploaded? how do i do this? linux mint btw