I use ffmpeg to:

  1. add multiple audio into a video file and offset them
  2. discard original audio in the video

Everything is OK except that 1. the audio volume in the result.mp4 is remarkably lower than the volumes in each individual input audio 2. sound quality in mp3(which is a background music) also drops

I listened to each 3gp and the mp3 files, can confirm that the two issues above.

Why is this happening?

ffmpeg -i 1.mp4 -i 1.3gp -i 2.3gp -i 1.mp3
  -filter_complex "[2]adelay=10000|10000[s2];[3:a][1:a][s2]amix=3[a]"
  -map 0:v -map "[a]" -c:v copy result.mp4
  • See stackoverflow.com/a/38714779/5726027
    – Gyan
    Commented Jul 24, 2017 at 13:09
  • Thanks, I was also checking that thread. The command there works better, but still it does generate an optimal result. For example, at the begin of 5 secs, there is only one mp3 file as audio input, but the volume is still very low. From a noob point of view, why can't ffmpeg just stack these audios up then generate result?
    – Xi Xiao
    Commented Jul 24, 2017 at 15:27
  • I checked your explanation @Mulvya in that thread. In my case, there is always two active audio inputs at any given time, one is the mp3, the other one is a 3gp, i.e. none 3gp files overlaps.
    – Xi Xiao
    Commented Jul 24, 2017 at 15:46


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.