-
-
Notifications
You must be signed in to change notification settings - Fork 7.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Specify time interval #2063
Comments
That is normal. ffmpeg can only cut the video at nearest keyframe
It is slower because we are using ffmpeg to do the download. There is no other option |
This isn't true, I regularly use FFmpeg to cut video accurately, it's perfectly possible to do when encoding the video (i.e. when not specifying |
Just remove the
|
Yeah my bad, you're right, I was confusing the seeking options even though I linked to the FFmpeg wiki page that explains them - seeking on input is the least accurate because it uses keyframes, not seeking on output. Seeking on output is what was necessary for accurate seeking. The problem is that when I seek on output with:
...I get the following error:
And an output clip that is time-accurate but with no audio. Any idea how this might be fixed? |
I can reproduce, but no idea why |
No longer reproducible |
Checklist
Question
I was wondering if there's a way to specify time interval to download, I have tried to use -ss and -to but it reported back with -t not an option (as noted at the bottom of readme, it's removed)
I have also found some info in issue #686, which did work for me, but somehow extracted with a slightly longer interval (included about 7 seconds before the specified time interval), and that also seems to be slower than typical download so I was wondering if there is another option (around 1-2% of full speed)
here's my line :
yt-dlp -f "bestaudio/best" --downloader ffmpeg --external-downloader-args "ffmpeg_i:-ss [time] -to [time]" --extract-audio --audio-quality 0 --audio-format opus [URL]
The text was updated successfully, but these errors were encountered: