#1151 A way to disable all audioprocessing for podcasting.


It would be very helpful to have an option to disable ALL audio processing, so that mumble can be used as a "cable" between to devices.
So we could use it for broadcasting and especially podcasting. So the interview partner can be patched in via mumble with studio audio quality.
I can clearly understand that mumble is primarily made for gamers, but it would be SOOO helpful for us podcasters to have this option. If i understand it right to do this you have to change some parts in the audioinput.cpp. It would be also very helpful to have this option in the iOS version too.

Udo Sauer
from sunny Wacken Germany


  • Sanya_Zol

    Sanya_Zol - 2014-04-22

    You can disable noise reduction and amplification by setting both sliders in "audio processing" to leftmost.

    Sometimes I'm using this to make things sound exactly how they was recieved from audio device (useful with virtual audio devices).

    Note that audio is strictly 1-channel and strictly encoded with lossy codec (opus or CELT - depends on version, server settings and connected clients) at bitrate <= 96 kbps. Also note that many settings (bitrate and framesize) settings applies only to outgoing transmission.

    • Dodge

      Dodge - 2014-09-28

      Even when I set Noise Reduction to Off, and Max Amplification to 1.0 (lowest setting), there is still some noise reduction happening. It is very clear when comparing a Mumble recording of my mic, to a Quicktime or Audacity recording of my mic, made at the exact same time. The Mumble version has less noise, but also a slight metallic/muffled sound, a clear sign of noise reduction.

      I'm also using Mumble for podcasting, and would much rather handle all my noise reduction in post, rather than introducing it into the source material.

      Edit: Nevermind user error. I confused artifacts from setting the Quality too low, to noise reduction artifacts.

      Edit 2: Nope. There is still some processing going on.

      Last edit: Dodge 2017-02-16
  • asdf asdf

    asdf asdf - 2016-03-13

    still a problem even in latest mumble.
    noise reduction just doesn't turn off all the way

  • Lyam Witherow

    Lyam Witherow - 2017-02-16

    Are you sure that this isn't a result of opus? If so, I would also like an option to tick to enable an extended slider. As an example 2.5 ms to 60 ms

    > Lowest       *                                   *             Highest
    > |------------|--O---------------|----------------|-------------------|
    > 2.5ms       5ms                20ms             40ms              60ms
    > Lowest (8kbps)                           *          (192 kbps) Highest
    > |----|----O----------|-------------------|---------------------------|
    >    Radio           Voice               Music

    The extended range can allow for someone to pick really crappy radio-like sound, or really great 192kbps sound.

    f doing this I would recommend a slightly more advanced feature in order to sync the audio a bit better, so that every client has delay compensation that syncs them up. This would mean that everyone would have to have artifical latency added to match whoever has the worst ping (under a certain threshold). So, the highest ping that is within 10ms would be the same ping given to anyone lower.

    This would be especially useful for same-room audio where one device may be on wifi with 40ms delay and other with 5 ms delay, wired.

    This second feature could be achieved by selecting two clients, and then selecting "Sync Clients"

    Last edit: Lyam Witherow 2017-02-16
  • Dodge

    Dodge - 2017-02-16
    Post awaiting moderation.

Log in to post a comment.