Recherche avancée

Sur d’autres sites (6976)

• How to convert audio in each directory by FFmpeg and batch script.

  16 juin 2013, par Stallman

  After a hard study, I have created a batch file :

  for /R  D:\storytelling\MusicFiles %%a in (*.3gp) do ffmpeg -i %%a -y %%~na.mp3

  However, it can't reach my goal since the scenario is a little different here :
  

  Here is my directory structure :
  

  I have several directories under a certain path :D:\storytelling\MusicFiles\ and new directories could be created by another application. And I put ffmpeg.exe file in this path : D:\storytelling\MusicFiles\.
  Also, In each directory, I have hundreds of .3gp files, and my target is to convert them to .mp3 in the directory where they used to stay at.
  But this script

  for /R  D:\storytelling\MusicFiles %%a in (*.3gp) do ffmpeg -i %%a -y %%~na.mp3

  would convert every .3gp file in each directory to the path : D:\storytelling\MusicFiles\

  It leads to :
  But I want the publicUser_XXX.mp3 files are still in the directory publicUser and after conversion all the files remain in their original directory. The only change is that I got new a media copy with different media format like :

  Please help and give me some advise.

• rtsp to rtmp using ffmpeg or any tool wrapper

  27 août 2017, par Chakri

  I have a requirement where I need to restream the RTSP stream from camera source to RTMP server. I know this may sound a repeated question but my exact scenario is I cannot do it manually over command line with ffmpeg command. I need a wrapper where I receive the rtsp and rtmp url from external source say through REST invocation. Then the code can trigger the ffmpeg restream.

  Basically flow is like this :

  1. Camera source application sends RTSP read event(could be basic HTTP(REST) request with RTSP url, metadata about camera info, serial no etc) to my streamer app

  Ex : /usr/bin/ffmpeg -i rtsp ://10.144.11.22:554/stream1 -f flv rtmp ://10.13.11.121:1935/stream1

  2. Streamer app computes the RTMP server url for corresponding camera and triggers a ffmpeg command to stream RTSP to RTMP

  3. Streamer app triggers above(2) in separate thread and keeps reading the logs for monitoring purpose. Also identifies the end of RTSP stream and sends an update(Example : RTSP END) event to UI

  Now at point(2) I need a suggestion. Here I need a stable wrapper/api which can help. I tried this through some Java wrappers but the process hangs or fails to read the output from ffmpeg. Also I need to handle streams from many cameras where spawning thread for each one could be exhaustive.

  So I am looking for some similar api/wrapper in C++ or Go Lang which might have more closer interaction in handling ffmpeg command.

  Please point if similar issue is addressed elsewhere

• Trim video file to edit-out silent parts

  18 décembre 2018, par karel

  I have video recordings of webinars with long silent gaps in them. I need to edit out the silent parts (both the audio and corresponding video tracks). I need to do this through a linux command and be able to do this as a batch action for several video files.
  Ideal scenario is to run a command that :

  1. detects long silent gaps (longer than defined time threshold, e.g. 5
     seconds)
  2. cuts out long silent sections and leaves only 5 seconds of
     silent video bet ween all ’loud’ video sections.

  In another post I found this great script (below). It does the job, but it a) doesn’t leave in extra silent 5 seconds of video which I need after every ’loud’ video part, and b) doesn’t compile produced video files into one composed video file (this problem is less of an issue for me, although it would be great to do that in the same command).

  I didn’t figure out how to modify the script to do this.

  Thanks beforehand for any suggestion.

  ffmpeg -i input.mkv -filter_complex "[0:a]silencedetect=n=-90dB:d=0.3[outa]" -map [outa] -f s16le -y /dev/null |& F=’-aq 70 -v warning’ perl -ne ’INIT $ss=0 ; $se=0 ; if (/silence_start : (\S+)/) $ss=$1 ; $ctr+=1 ; printf "ffmpeg -nostdin -i input.mkv -ss %f -t %f $ENVF -y %03d.mkv\n", $se, ($ss-$se), $ctr ; if (/silence_end : (\S+)/) $se=$1 ; END printf "ffmpeg -nostdin -i input.mkv -ss %f $ENVF -y %03d.mkv\n", $se, $ctr+1 ; ’ | bash -x