Hello Tim, thank you for your guidance! I indeed didn't account for the number of channels or samples when calculating num_samples and the duration. I also wasn't totally clear with myself about whether I was counting samples or frames (ie. one sample for each channel), which may have also been causing me problems. The program can now successfully play mono 32-bit float samples (<a href="http://gstreamer-devel.966125.n4.nabble.com/file/t378119/basic-tutorial-8s.bak" target="_top" rel="nofollow" link="external">this is my updated code</a>), however when I give it a stereo sample, it seems to be played at half the speed and seems to be drawn twice on the wavescope, despite me setting the number of channels to 2 in the code: <img src="http://gstreamer-devel.966125.n4.nabble.com/file/t378119/03.png" border="0"/> I imagine that a possible cause for this could be that multiple channels are stored differently in the file than GStreamer expects. I think that the sample data is stored in the file with the channels alternating every sample -- is this how GStreamer expects the samples to be stored? The program now also sends an EOS event once the end of the sample is reached to prevent a seg-fault: <pre> if (data->playback_offset >= data->size) { gst_app_src_end_of_stream ((GstAppSrc *) data->app_source); } </pre> This works, however the program plays silence for once more the sample's length after the end is reached. Do you have any idea about what may be causing these problems? Thanks -- Albert <hr align="left" width="300" /> Sent from the <a href="http://gstreamer-devel.966125.n4.nabble.com/">GStreamer-devel mailing list archive</a> at Nabble.com.