Best recording settings

Hi,



I’m trying to use Core v2 to record a conversation for playback, and I’m trying to get the best recording quality possible out of the device. Currently the recording is quite noisy and I was wondering how to improve this? I’ve tried the following:


  1. Recording directly off the sound device:
    </s>arecord -c 2 -r 48000 -f S16_LE output.wav<e>

  2. Getting the raw messages from the respeakerd socket:

    respeakerd_source.py:117
    </s><i> </i>if decoded_data: recording_file.write(decoded_data) demo_respeaker_v2_vep_alexa_with_light.py <e>

Both of these produce roughly the same result. Do I need to interface directly with librespeaker to get noise supression like in <URL url="https://forum7.seeedstudio.com/t/noise-suppression-with-librespeaker-in-respeakercorev2/6089/1 text=“viewtopic.php?f=89&t=8196”>https://forum7.seeedstudio.com/t/noise-suppression-with-librespeaker-in-respeakercorev2/6089/1


Thanks