Unfortunately, our mpeg demux is having problems parsing multiple audio streams out of a file and selecting the appropriate stream to play. We will enable the parsing of multiple audio streams in the future.
There is a work around you can use today. If you have access to the content, you can strip the excess streams using a transcoding tool like ffmpeg:
% ffmpeg -i MyMultiTracAudio.m4a
....
Duration: 00:59:06.98, start: 0.000000, bitrate: 71 kb/s
Stream #0.0(eng): Audio: aac, 44100 Hz, stereo, s16
Stream #0.1(eng): Data: mp4s / 0x7334706D
Stream #0.2(eng): Data: mp4s / 0x7334706D
Stream #0.3(eng): Data: rtp / 0x20707472
% ffmpeg -i MyMultitracAudio.m4a -map 0:0 MySingletracAudio.mp3
% ffmpeg -i MySingletracAudio.mp3
....
Input #0, mp3, from 'MySingletracAudio.mp3':
Duration: 00:59:07.92, start: 0.000000, bitrate: 63 kb/s
Stream #0.0: Audio: mp3, 44100 Hz, stereo, s16, 64 kb/s
The -map option can map input streams to output streams. In this case I also transcoded the stream from aac to mp3.
By default on Fedora, ffmpeg is not compiled against libfaac so the transcoding was a necessary step for me (libfaac is required for encoding aac).
--Kevin