Since I couldn't find one I tried to write my own tool in this stackoverflow question.Received command: -i input.webm -i input.webm -filter_complex ' concat=n=2:v=1:a=1 ' -map -map output.mp4.

Thus this certainly is not a global solution. This may result in incorrect timestamps in the output file.Īnd then I would hear a very faint 'gap' between two files.

Sometimes (not sure what triggers this), ffmpeg would give the warning: Non-monotonous DTS in output stream 0:0 previous: 5011328, current: 5011200 changing to 5011329. So I guess really no re-encoding happens.īut this does not seem to work on all input files. If I look at the raw hexdump of my input files, then I can find the vorbis packets exactly represented in the output file. $ ffmpeg -f concat -i inputs.txt -c copy out.ogg It seems that ffmpeg can sometimes do it using the concat demuxer: $ cat inputs.txt If that would be okay, then I could keep the original ogg files and would use less space. I also want to avoid concatenating the files into a lossless format like flac as this would unnecessarily blow up the file size. That method also does not work if the input ogg files happen to all have the same stream id.Īdditionally, since I'd like to throw away the original files and only keep the concatenated version, I'd like the concatenation to be done lossless, just as the cat solution.

Other programs which don't understand it are gstreamer. I know that the ogg vorbis format supports plain concatenation of multiple files like: $ cat file1.ogg file2.ogg > output.oggīut sadly not all players are able to understand files created like that, for example my mobile audio player and I'd avoid to buy a new one. I have a couple of ogg vorbis files, all encoded with the exact same properties that I want to concatenate into a single file.

Ffmpeg concat vorbis encoder error