In recording of a SNS chat, voice from MIC is very low

I tried to make a recording of a SNS chat, voice inputs come from two parties: (1) SNS app (chatting counterpart), and (2) MIC (me), however, the vioce of my part in the recorded chat is very low (still able to hear), the chatting counterpart is normal.

  • Checked recording related settings on either Audacity or system, all slid to max.
  • Under a MIC setting related menu of Audacity: among following 3 options:
    “Windows Sound”
    “MIC Realtek High Definition”
    “Stereo Mixer”
    the 3rd was chosen.
  • However, if changing to any of other two above, MIC recording level will be normal but the counterpart
    will turn very low!
  • MIC tested with Skype, it seems fine and normal.
  • Audacity 2.1.2
  • Windows 7 Home Premium (x64)

Please see attached screenshots for reference.

What kind of issue could it be? an issue with Audacity? is there any way to fix / workaround?

Audacity is not recommended for recording Chat, Skype, Conferencing or Game Commentary.

Audacity can’t deal well with multiple sound “devices.” The local microphone is typically one device and the far side is a separate one. Getting a good balance is really unusual.

You have an unusual condition. Usually the local microphone works just fine and nobody can hear the far side.

You should find out if the app has a forum and if they have a recommendation for capture.

The more serious podcasters and broadcasters…and me…all use two computers and a mixer; or a Chat computer, small mixer and sound recorder. It’s when you put the recorder on the same computer as the Chat you run into problems.



Thanks for your prompt response and kind full-length explanation!
Well understood.

You said Audacity is not supposed for use with chat recording situation, not dealing well with multiple sound devices… while my case seems under some special conditions which makes it work though not perfectly.

From your comment, it implies Audacity has the potential supporting multiple sound devices, then why not by investing some further effort to make it an explicitly supported, new feature?

Btw, attaching the screenshots forgot to last time.

It doesn’t. Audacity only supports recording from one device at a time.
If that one device supports multiple channels, then Audacity can record up to the number of channels that the device has, but only one device at a time.

The “device” does not need to be a physical (hardware) device. Audacity can also record from virtual devices (provided that they are supported by PortAudio). The distributed version of Audacity does not support ASIO due to licensing issues. Virtual devices (such as “Virtual Audio Cable” or “Voicemeeter”) may provide a way to record multiple devices, though please note that these are third party products (not made by us).

Audacity can seem to be recording from both sides when what it’s really doing is recording one thing very well and faulty echo cancellation of the other. A variation of this is a complaint that the far side is gargly or bubbly and how do they clear it up. You usually don’t. It’s a cousin problem of compression distortion. There’s pieces of sound missing and non-recoverable

Every so often, someone will discover that they can record, for example, Skype perfectly and create a production around it. Sometimes they post notes how easy it is and we get a flurry of postings how hard it is. Those successful people are celebrity unicorns and what they’re doing is very unusual.

I’m guessing they’re taking advantage of an anomaly or defective service in their computer. If they change or upgrade to a different machine, it will stop working.

In general, we don’t get complaints about people using purpose-built recording software (there have been problems with those, too) and no complaints from people using the two computer and mixer technique.


If the two conversations are relatively clean and not too far apart, you can try LevelSpeech

or Chris’s Compressor.