I'm using the OPUS to stream my audio, but the quality is very pour. I'm using this pipline for RX: gst-launch-1.0 udpsrc port=9001 caps=\"application/x-rtp,rate=48000,channels=2\" name=teste ! rtpopusdepay ! opusdec ! level message=true ! tee name=t ! queue ! audioresample ! alsasink t. ! queue ! lamemp3enc and this for my TX: gst-launch-1.0 autoaudiosrc ! level message=true ! audioconvert ! audioresample ! tee name=split ! queue ! opusenc bitrate=48000 bandwidth=narrowband complexity=0 inband-fec=true packet-loss-percentage=10 ! rtpopuspay pt=100 ! udpsink host=192.168.10.142 port=9001 name=teste split. ! queue ! lamemp3enc Does anyone know how can I improve my audio?
What is your hardware platform and which BSP layer are you using? Is the hardware capable of handling the two pipelines that you present in an older release or are you simply asking too much (performance and no latency) for this board? Have you considered adjusting the priority of the process if they are intended to be real-time? That might help.
Without additional information, there's no action to take so defect closed.