Found it ! Forward microphone to secondary line-out


Ok, this is cool. Except for energy bill.
I found finaly how to forward the HTC vive microphone to external line-out of the sound card. So in the future i can route everything first external and do the required things to the signal or just have a better feeling that sound is off. I need to dig a bit deeper. but the trick is in here.

  • Right-click on the volume icon and select Recording Devices
  • Double-click on your microphone
  • Select listen tab
  • Check the "listen to this device" box

Now i need a few small modifications. extra D/A converter to get it complete working.


When you use this listen option, it is not latency optimized. There will be a one full buffer delay in the pass-through. Expect approximately 0.15s delays.


Still beter then always have the wrong audio settings,.
Microphoens enabled when you not want it
And more drama like that.

Biggest design flaw with HMD. own soundcard build in.
And no support for line-in and line-out on the box.

But for now i still use the wrong way with direct HMD audio use.
Besides. i prefer text chat.

Are you anyway sure there’s a delay like that ? Depends mabye on the hardware to.

Besides, the delay does not have any effect i think. because the moth etc. only respond at the moment high fidelity get audio. and that is after it returned to the line-in. SO the dealy must not be a problem at all.


Yes, there is a delay using the listen option. It is meant as a test/diagnostic pass-through. The Windows audio core is very low latency. It uses audio buffer cursor (which means the software knows where the audio hardware is reading/writing). That is low latency. But to do that you need to use the programming interfaces.

BTW, I do understand why low latency I/O points are not available in the interface app. The core audio systems of Windows, OS X and Linux (and the myriad linux variants) are non-trivially different from each other. Getting a common abstraction layer that itself does not add latency is difficult. It’s not impossible to do, but I can imagine the difficulties making all that work reliably throughout the development process for all the supported platforms.


But i still think that somoneone inworld would not notice the latency. because the Signal in my future setup goes out on 1 card. Get true soem aurio hardware and then back line-in on other card. And from there it go to high fidelity.

SO moth and voice still need to be in synch. Only hand animation can be oof be 0.15s You mean 150ms right ? or mabye lower. Anyway, nobody will notice that your hand movement is not in sync with voice i think.


Yes, 150ms. That latency is similar to European to U.S. connection. So, if you are in Europe then it would be a total of 300ms, more or less.