VB-Audio's Forums

Hello everyone.
(I'm sorry for my English, not my native language)
I have quite a big problem with voicemeeter and I don't know how to fix it.
Recently I've upgraded my PC and switched Intel to amd (not sure is there any chance that this is because of that)
Before I had 7700k at z270 tomahawk board, now I have 3700x with asus prime x570 pro mb.
On intel , after setting everything, EQ for mic work perfectly, now it's ok for few sec and than my voice is cracking/lagging.
Mic is fine (Blue Snowball ICE) it's something with voicemeeter/vb-cable or something else.
here is a vid to show how it looks: https://www.youtube.com/watch?v=UzBYcniAieM
I've saved settings for VM before "format" than just imported it and thought that sound will be fine (well it was for a while) but than during stream somebody told me that something is wrong with my voice.
Switching to just mic without EQ and there is no problem at all, once i pick a VM, its unable to understand what I'm saying.
I use same vb cable and vm, first I've installed cable tham vm, reinstalled cable already and still same issue.
If something is hard to understand, pleas tell me and I'll try to explain it as good as I can.
Thank you in advance for any reply.

It kinda sounds like what sometimes happens for me with Cantibile when the Insert doesn't work correctly. The absolute worst case is normally restarting and see if it works.

However, that is normally if it works in most times. On the basic of the video, you're diagnosing the check on OBS, and yes, that is good that you get the audio to show it works right when using straight up the yeti and its only when you connect to whatever your VM output is for that mic that it becomes an issue. So to me, since it can work just fine, you're right to believe it could be a voicemeeter issue.

But my question, why are you setting the output of A1 to the VM Audio Cable? While I partially don't think it could be the issue, it is possible that its trying to handle a potential triple buffer unintentionally. I know I use Potato, but the A1 effect I believe should still apply. I am guessing that the desktop sound on your Arctis 7 Game headset (seen on OBS) is supposed to be your regular speaker? If so, try setting that to A1, restart the audio engine, and test again in OBS and make sure the Desktop Audio is set to Voicemeeter's VAIO and not the Arctis headset.

Hopefully you understand this.

Hello.
Thank you for reply.
Basicaly i set everything like in this video: https://youtu.be/S9vznpXS_so?t=260
Hardware input-snwowball, hardware input 2 (xonar DX, my sound card I use paired with Arctis 7), virtual input is VB adio VM VAIO and as hardware out A1 WDM Cable input.
Before when I had all like this, I had no issues at all, now my voice is cracking after few sec. What I tried now, instead of WDM Cable input, i set A1 to MME Cable input, was fine but can't tell for how long.

Hmm. So just to get this straight, you have 2 sound outputs? Your headset and your soundcard?

If so, I think it would be better to make A1 whichever is the stable sound (I guess in this case, your soundcard more than likely) and A2 your headset.

Using this pic as an example of my setup...
I set my Headset Mic as Hardware Input 1. Its audio is fed into VAIO Aux (B2)
I set my onboard Audio Mic port to Hardware Input 2. Its Audio is fed into VAIO (B1)
I used VB Audio Cable to Hardware Input 3 so I can stream music from Pretzel to that (and roughly any music source really). Its audio is fed to A2 (which is my headset) just for me to hear, but in OBS, its also played there directly.
Ignore the Hardware Input 4 section for now cause I'm working on that.

On VAIO 3, you notice I labeled SFX Out. Currently, I have SFX fed into VAIO 3, then have it played out to A1, A2, B2, and B3. B2 is the VAIO Aux so if I'm on Discord, people hear the SFX as well. B3 is for OBS to have the SFX played as well.

In your case, to me, it seems like you'd want either Voicemeeter Banana for I assume more control output audio or you're looking for something else in regards to the Voicemeeter Cable to combine 2 audio sources that doesn't sound like it should be routed that way, but I guess you could.

Also Settings. I don't know if the buffer is something you want to mess with because in my case, it works, but if I'm going by Vincent, I'm probably pushing it for lowering the ms response rate, but that was my intent. You're at 512 from what I see so you're fine there because we don't want to change that until we isolate the issue as you should have more than enough buffer room.

Hello.
Thx for reply.
Just a quick update.
If I set in voicemeeter as hardware input (1) snowball, than as A1 my sound card (Xonar DX-Speakers) than I hear myself in headphones all the time and only me, no other sound.
Everything else goes through speakers in my 2nd monitor (hdmi-nvidia)
Tbh, I don't know much about this software, actually nothing at all, just found a vid about how to make my voice better with EQ, tried with virtual cable+voicemeeter and result was quite amazing for me.
Now after fresh win 10 install after upgrading my pc, same settings doesn't work.
For more detailed information how everything is setup on my pc.
Snowball is my main and only mic.
Xonar DX is connected via aux cable (front speaker output) into Arctis 7 transmitter (Line in)
I don't need SteelSeries software and sound from xonar is simply better even if it's still wireless.

Just and update.
Curently my A1 device is set to MME: Cable input (VB audio virtual) and it works ok.
Don't know what's the difference between WDM and MME but it works fine now.

Did do a quick contact on ya but I didn't see anything so I guessed you didn't get it.

WDM has less response time after process and utilizes the API that Windows has had since Windows Vista.

MME should really at this point be used when WDM isn't working which is older and offers more latency between the sound output. A tool will show you the average time, but even using a Karaoke tool where it plays the audio and you singing, you'll notice the difference between when you'd hit the note to where the audio processes it. So if you went live on stream with a camera, people might notice your speech a little delayed from your mouth movement.

VoiceMeeter at its basic is designed very well to do the kind of sound that sounds like you're trying to do.
Just to compare with the earlier pics, you'll see that I literally have everything handle the sounds proper when going into OBS. That is also with the expansion when I went into Potato.

That said, I think you may need to switch to Banana only because that if you're hearing yourself and looking at the settings, you may not have that much more option since looking at it on my end, its all 1 thing for less confusion. Banana would be a little closer to my pic, but with 2 outs, and you can send your mic to your speaker IF YOU WANT, but it can just go out as normal.

Just to be sure, when you did the test, you didn't have OBS on, right? If so, maybe when you heard yourself, it was the audio monitor that was on?