I have not been following the issues with ffmpeg and avconv forking, but the ubuntu build seems to have deprecated ffmpeg in favor of avconv. So I decided to post the settings I am using for avconv in case anyone is having conversion issues.
Although the avconv syntax is almost the same as ffmpeg there are few changes (changed bitrate from -ab to -b, the quality from -v to -q. The default output for avconv seemed pretty verbose, so I added -loglevel error):
New Subsonic Settings:
Name: mp3 audio
Convert From: ogg oga aac m4a flac wav wma aif aiff ape mpc shn
Convert To: mp3
Step 1: avconv -i %s -b %bk -q 0 -loglevel error -f mp3 -
Step 2 is blank since its not needed.
Here is an explanation of the switches to help anyone understand:
-i %s (take the input from the file stream %s)
-b %bk (set the bitrate to the variable %b and add a k since the bitrate is in thousands, ie "128k")
-q 0 (set the quality to the best, this value varies from 0 to 31, with 31 being the worst. Could lower if your machine processor sucked)
-loglevel error (set the output logging level to errors only)
-f mp3 (set the output format to mp3)
- (direct the output to the pipe so that it can be streamed to the player)
NOTE: I noticed m4a transcoding was not working and was giving the error "Encoder (coded id 0) not found for output stream #0:0" error message in the log. This was due to the upgraded Ubuntu not having all the transcoders by default (because they are restricted by law without a license). In order to install those transcoders, one should install the "ubuntu-restricted-extras" package in the Ubuntu Software Center. This will load transcoders needed to do many of the audio format conversions you might have.