OBS reduces microphone audio

  • Thread starter Deleted member 427361
  • Start date
D

Deleted member 427361

When exporting a video, having the volume at 2000% or 100%, the audio is barely heard, only with OBS, if I try other programs to record audio my volume is normal.
 

AaronD

Active Member
What other programs? Do they have some audio processing built-in, that you have to add explicitly in OBS?



I assume you're recording a mic, and expecting a similar volume from the speakers as a commercial release or TV show. You can't just send it straight across and get that. The reason has to do with "live" vs. "mastered" signals.

A live signal, like what comes from the mic, is unpredictable in what it's going to do next. You don't want a sudden jump in volume to be distorted, so you need to keep some "headroom", or unused volume, for those sudden jumps to work in. How much headroom is anyone's guess, but a general rule in professional live work is about 18dB. That's kind of a lot, but we often do end up using it.

A mastered signal, like a TV show, has already tamed everything and knows exactly what's coming. So its goal is to minimize noise in the cheap transmission medium. (TV channel, radio, CD, cassette tape, etc.) To do that, they simply turn it up to use all the headroom it has, and expect the end-listener to set it where they like it. This keeps the end-listener's volume setting as low as possible, which effectively turns down the noise.

A live broadcast is both. The input is live, and the output is mastered. So you need to have something in between to make that conversion. A recording app that tries to hide all the complexity, might have some automatic processing that *usually* works for that, but OBS doesn't. On purpose. Not automatic anyway, because it has no way to know that "this is a mic" and "that's an external mixing console".

You can, however, add a Compressor filter to what you know to be the mic (maybe two Compressors with different settings, or a Compressor and a Limiter), and tweak them by ear (or have someone else tweak them by ear while you feed the mic) to knock the peaks down without sounding "processed" and put the result up to where a mastered broadcast is expected to be.
 
Top