Problem Recording Stereo Mix following Windows Update

There was a Windows 10 update several months ago, which ruined the Stereo Mix. I didn’t notice fast enough to be able to roll back the update.

Have manage to re-install the Stereo Mix and some research showed that you have to allow apps access to the microphone (even if you don’t have one?) for it to work at all, but I am not getting the previous quality as before. I record via the Windows WASAPI… this worked perfectly prior to the update. If I recorded a song that had been hard limited in mastering, this would recreate that perfectly which showed that I was getting a pretty much perfect replication of the original audio. Since the update, there is now a 20 or 21 khz cut off and looking at the hard limiting shows degradation in the sound because of this. Anyone know how to fix this?

Thanks.

Try setting stereo-mix & Audacity project rate at 48kHz
(Bear in mind most people cannot hear above 15kHz, so anything above 20kHz is irrelevant).
stereo-mix properties.png

Thanks, but that didn’t do anything. I want to record stuff losslessly without any kind of downsampling, which I did with no issue before the Windows update last year. There is presumably some playback issue, but cannot figure it out.

There is a process called dithering which adds noise, which is mostly visible on the high end of the spectrogram …

dither noise.png
You can switch dithering off in Audacity, (but supposedly that makes it sound worse) …
https://manual.audacityteam.org/man/dither.html

I assume you are using WASAPI loopback to record streaming audio?

I believe WASAPI loopback is at least as “accurate” as stereo mix. Stereo mix is a driver feature and it’s becoming rare. Your drivers may have been updated.

Since the update, there is now a 20 or 21 khz cut off and looking at the hard limiting shows degradation in the sound because of this.

I doubt that’s the problem. Even if you can hear beyond the normal human hearing range in a hearing test, those super-high harmonics are drowned-out in normal music.

And how do you know the streaming audio goes beyond 20kHz?

Maybe Windows or a driver utility is applying some “enhancements” and altering/degrading the sound? Still, you should record what you’re hearing but if you play back the recording those alterations would be applied again.

If I recorded a song that had been hard limited in mastering, this would recreate that perfectly which showed that I was getting a pretty much perfect replication of the original audio.

What? And, where are you getting the “original”? Most streaming services use lossy compression which might “hide” hard-limiting or clipping, and most of these services also apply “loudness normalization”.

I want to record stuff losslessly without any kind of downsampling,

How do you know the “original” streaming sample rate? Up-sampling or down-sampling is usually transparent as long as you stay above “CD quality” (44.1kHz, 16-bit). Or you can set Audacity to 96kHz and up-sample as you record.

To turn off Windows “enhancements”, you can do the following:

  1. In the Windows search bar, type: “mmsys.cpl”
  2. On the window that pops up, click on the “Recording” tab.
  3. Click on “Stereo Mix”
  4. Click “Properties”
  5. On the next window that pops up, click on “Enhancements”
  6. Then select “Disable all sound effects”

I hope this helps. :smiley: