avconv -y -i input.avi -b 915k -an -f mp4 -ar 44100 -f s16le -ac 2 -i /dev/zero -acodec libfaac -ab 128k -strict experimental -shortest -vcodec libx264 output.mp4 -loglevel fatal
First of all, this seems to be an old version of avconv, since the command line has changed since then (but not too much).
So, let's break it down:
-y
This answers 'yes' to questions like "do you want to overwrite the output file".
-i input.avi
This gives the program the file input.avi as an input
-b 915k
This asks to change the bitrate to 915 Kibibytes per second
-an
This removes all the audio from the output.
-f mp4
Sets up MP4 as the format of the output file
-ar 44100
This sets audio sampling rate of the following input file.
-f s16le
This sets the format of the audio of the following input file.
-ac 2
This sets number of channels of audio to two.
-i /dev/zero
This adds another input file that consists entirely of zero input
-acodec libfaac
This reencodes the audio (silence most likely) with libfaac
-ab 128k
Setting the audio bitrate to 128 Kbps
-strict experimental
Allows avconv to use nonstandard approaches while encoding.
-shortest
Ends encoding when the shortest of the inputs has ended. This is needed because /dev/zero will never end.
-vcodec libx264
This sets the library to do the video encoding. The codec will be (unfortunately) h264
output.mp4
This is the name of the output file
-loglevel fatal
Fatal messages will be written as the log, and that's it.
In the future you may find man avconv to be your friend.
Related
I have one image and one video. I would like to add image as video just before actual video so i can stream video with intro frame for 5 seconds.
I found one command and full fill my requirement but in that command image added at the end of the video and i need beginning of the video. Let me share command below:
ffmpeg -i 1.mp4 -loop 1 -t 5 -i 2.jpg -f lavfi -t 5 -i anullsrc -filter_complex "[0]split[base][full];[base]trim=0:5,drawbox=t=fill[base];[1][base]scale2ref=iw:ih:force_original_aspect_ratio=decrease:flags=spline[2nd][base];[base][2nd]overlay='(W-w)/2':'(H-h)/2'[padded];[full][0:a][padded][2:a]concat=n=2:v=1:a=1[v][a]" -c:v libx264 -c:a aac -strict -2 -map "[v]" -map "[a]" output.mp4
Image should be resize dynamically according to the video resolution.
Best solution will be appreciate from bottom of heart.
ffmpeg -i 1.mp4 -loop 1 -t 5 -i 2.jpg -f lavfi -t 5 -i anullsrc -filter_complex "[0:v]trim=0:5,drawbox=t=fill[base];[1][base]scale2ref=iw:ih:force_original_aspect_ratio=decrease:flags=spline[2nd][base2];[base2][2nd]overlay='(W-w)/2':'(H-h)/2'[padded];[padded][2:a][0:v][0:a]concat=n=2:v=1:a=1[v][a]" -c:v libx264 -c:a aac -map "[v]" -map "[a]" output.mp4
No need for the split filter.
Do not re-use labels. Each output label must be unique. For example, you used [base] several times. So I renamed the next one [base2].
Order of video is determined by the order given to the concat filter. I re-arranged it so [padded][2:a] plays before [0:v][0:a].
-strict -2 hasn't been needed since 2015 (it was for the AAC encoder). You don't need that unless your FFmpeg is very old.
I'm trying to stream a webpage captured with PhantomJS to Youtube using FMMpeg.
This is the command I use:
xvfb-run phantomjs --web-security=no render.js | ffmpeg -threads 0 -y -v verbose -c:v png -r 30 -f image2pipe -i - -f lavfi -i anullsrc -strict -2 -acodec aac -ac 1 -ar 44100 -b:a 128k -c:v libx264 -s 1280x720 -pix_fmt yuv420p -f flv "rtmp://a.rtmp.youtube.com/live2/key";
And the render.js code:
http://pastebin.com/raw/X9gv8iGH
It looks like it's streaming, but no feed is received by YouTube, and I can't see where the problem is.
Outpout from my console
Try this:
phantomjs --web-security=no render.js | ffmpeg -threads 0 -y -v verbose -c:v png -framerate 33 -f image2pipe -i - -f lavfi -i anullsrc -strict -2 -acodec aac -ac 1 -ar 44100 -b:a 128k -c:v libx264 -s 1280x720 -pix_fmt yuv420p -g 60 -r 30 -f flv "rtmp://a.rtmp.youtube.com/live2/key";
Parameter -framerate:
You can specify two frame rates: input and output.
Set input frame rate with the -framerate input option (before -i). The default for reading inputs is -framerate 25 which will be set if
no -framerate is specified.
The output frame rate for the video stream by setting -r after -i or by using the fps filter.
So in your case framerate should be 1/period_from_phantomjs which is 1000/30 = 33.33
As for the -g 60, that will add a key frame every 2 seconds, which is probably a requirement for the youtube streaming api (I know that for facebook it is).
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Closed 6 years ago.
Improve this question
I have two IP cam and a computer with ffmpeg to re-stream them to YouTube.
First Cam
It has Video and Audio. I can re-stream to YouTube successfully with this command.
ffmpeg -re -rtsp_transport tcp -i rtsp://192.168.aaa.bbb:554/ch0_1.h264 -acodec libmp3lame -ar 44100 -b:a 128k -pix_fmt yuv420p -profile:v baseline -s 426x240 -bufsize 2048k -vb 400k -maxrate 800k -deinterlace -vcodec libx264 -preset medium -g 30 -r 30 -f flv rtmp://a.rtmp.youtube.com/live2/<YouTube_key>
Second Cam
It has Video only. However, with both commands below, I can't stream to YouTube. (although the ffmpeg process is running well)
ffmpeg -re -rtsp_transport tcp -i rtsp://192.168.aaa.ccc/Streaming/Channels/102 -an -pix_fmt yuv420p -profile:v baseline -s 426x240 -bufsize 2048k -vb 400k -maxrate 800k -deinterlace -vcodec libx264 -preset medium -g 30 -r 30 -f flv rtmp://a.rtmp.youtube.com/live2/<YouTube_key>
or
ffmpeg -re -rtsp_transport tcp -i rtsp://192.168.aaa.ccc/Streaming/Channels/102 -acodec libmp3lame -ar 44100 -b:a 128k -pix_fmt yuv420p -profile:v baseline -s 426x240 -bufsize 2048k -vb 400k -maxrate 800k -deinterlace -vcodec libx264 -preset medium -g 30 -r 30 -f flv rtmp://a.rtmp.youtube.com/live2/<YouTube_key>
I've test both camera's stream first using FLV, and they work well. So, the issue is not the URL of IP camera.
From what I understand, YouTube define stream as audio-video content. However, what if the source only has Video.
So, my question:
How to live stream YouTube without audio, specially from an IP cam?
Thanks
Generate a dummy audio and try
ffmpeg -re -rtsp_transport tcp -i rtsp://192.168.aaa.ccc/Streaming/Channels/102
-f lavfi -i anullsrc
-pix_fmt yuv420p -profile:v baseline -s 426x240 -bufsize 2048k -vb 400k -maxrate 800k
-deinterlace -vcodec libx264 -preset medium -g 30 -r 30
-acodec libmp3lame -ar 44100 -b:a 128k
-f flv rtmp://a.rtmp.youtube.com/live2/<YouTube_key>
I have a problem that to save Output RTP as a file.
(Is that a possible? Am I Right?)
Trans-coding goal as below:
1. Save the RTP stream to file in local storage using FFMPEG.
2. Input is file.
3. Output is RTP stream file.
I`m using that.
./ffmpeg -re -i ../Video_Sample/03.Fashion_DivX720p_ASP_87s_1000k_720p.mp4 -c:v libx264 -b:v 1000k -preset superfast -an -f rtp -y test.rtp
But I got a message like that :
Could not write header for output file #0 (incorrect codec parameters ?): Invalid argument
How can I fix it?
RTP is the Real-time Transport Protocol and not a file.
If you want to stream your mp4 file you could do it as followed:
ffmpeg -re -f mp4 -i ../Video_Sample/03.Fashion_DivX720p_ASP_87s_1000k_720p.mp4 -vcodec libx264 -b 1000k -preset superfast -an -f rtp rtp://hostadress:port
Did you mean rpt (Report File) file?
I have troubles with setting constant output bitrate. Every time I try I am getting same bitrate as input (I wanna downrate the file)
As input I have MPEG2/MPEG-A file.mpg with VBR 10Mb
As output I want MPEG2/MPEG-A file.ts with CBR 8Mb
avconv -i file.mpg -codec copy -b 8M -maxrate 8M -minrate 8M -bufsize 4M -f mpegts file.ts
Is there something wrong there? Can you suggest me better parameters so I can get better output quality?
-codec copy tells avconv to just copy the first video, audio and subtitle bitstream of the input to the output. There will be no re-encoding, so any of -b, -maxrate, -minrate or -bufsize don't make sense.
So:
Remove -codec copy.
Use -b:v instead of -b since -b alone is ambiguous and could refer to both video and audio.
Copy the audio stream with -c:a copy.
In essence:
avconv -i file.mpg -b:v 8M -maxrate 8M -minrate 8M -bufsize 4M -c:a copy -f mpegts file.ts