edsu · May 5, 2016 12:14
diff --git a/combine-tracks.sh b/combine-tracks.sh
 # usage: combine-tracks.sh input.mov output.mov
 #
 # I've been using this ffmpeg command to take recordings made with ecamm's Call Recorder for Skype and uploading them 
 # to YouTube for transcription. I found that YouTube would only transcribe one of the audio channels which
 # was often my own voice, which made it less than useful, to put it mildly. Once the channels are combined YouTube will 
 # transcribe both sides of the conversation. I also found that it was useful to adjust the volumes of the tracks.

 ffmpeg -i $1 -filter_complex "[0:a:0]volume=0.781250[a1];[0:a:1]volume=0.781250[a2]; [a1][a2]amerge[out]" -map 0:v -map "[out]" -c:v copy -shortest $2
	# usage: combine-tracks.sh input.mov output.mov
	#
	# I've been using this ffmpeg command to take recordings made with ecamm's Call Recorder for Skype and uploading them
	# to YouTube for transcription. I found that YouTube would only transcribe one of the audio channels which
	# was often my own voice, which made it less than useful, to put it mildly. Once the channels are combined YouTube will
	# transcribe both sides of the conversation. I also found that it was useful to adjust the volumes of the tracks.

	ffmpeg -i $1 -filter_complex "[0:a:0]volume=0.781250[a1];[0:a:1]volume=0.781250[a2]; [a1][a2]amerge[out]" -map 0:v -map "[out]" -c:v copy -shortest $2