I have a modem which accepts a voice in GNU Radio, digitizes and compresses it (codec2) and transmits it using USRP on one laptop and receives it on another laptop. When I use "Wave File Source" in the transmitter everything is OK and I clearly hears the voice from "Audio Sink" at the receiver, but as I switch to "Audio Source" at the transmitter, nothing can be heared! On one laptop, there is no problem but with sample rate of 1.6 MHz, there is about 3 seconds of delay in voice.

I wonder why and would be pleased if someone introduces some tricks in working with "Audio Source/Sink" in GNU Radio.

  • $\begingroup$ do you happen to get anything like U printed on your console? Both your USRP and your sound card have a physical sampling rate, and the processing between these two must, in effect, convert the input rate to the output rate. Note that neither sound cards nor USRPs can run at arbitrary sampling rates! $\endgroup$ Jan 8, 2022 at 12:14
  • $\begingroup$ It shows some "U" at the Tx but stops after a while (not continuously printed) $\endgroup$
    – Ali
    Jan 8, 2022 at 12:19
  • $\begingroup$ hm, that doesn't sound good. Does it by any chance tell you it's stopped doing, or emit another single letter like "S" or "D" before stopping? Is some == 2 or is some more like 200? $\endgroup$ Jan 8, 2022 at 12:25
  • $\begingroup$ let's ask this explicitly: does your signal processing make sure that the 44.1 kHz, or 48 kHz, or whatever your audio source produces is converted to the sampling rate of your USRP sink? $\endgroup$ Jan 8, 2022 at 12:26
  • $\begingroup$ Maybe add your GNU Radio flow graph to your question? If you're using GNU Radio companion (GRC), there's "File-> Screen Capture", which will be very useful. $\endgroup$ Jan 8, 2022 at 12:27


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.