New Audio Equipment

A couple weeks ago I moved my microphone arm and ever since my Bue Yeti mic has been having an identity crisis. This has resulted in it appearing as a generic USB Audio Device in multiple machines. This is already my second of this microphone and I don’t want another one that will break if I move the arm it’s on in the wrong direction. While it still technically works the quality of sound has taken a major hit and it doesn’t work quite the same way every time I go to use it.

I’ve been researching what my next microphone would be for a little while ahead of this incident. There are a number of high quality USB mics that I considered but ultimately decided that not to go in that direction. From in my audio rabithole I’ve determined that the best sound quality would come from an XLR style instead of USB. Lots of people suggest the SM7b or the Procaster.

I ended up going with a SM57 by Shure. This microphone is one of the best selling in the world and has been used for presidential speeches since it came out in 1965. For me it works well due to its small size and smaller price tag.

Moving from the Yeti which is a condenser mic to the dynamic SM57 is going to require me to use it a bit differently. For instance it will need to be closer, probably in the video if I’m on a call, and to speak up. Since dynamic microphones are less sensitive room noise should be less of an issue but if I don’t speak louder and directly into it my voice will be quiet and flatter than it may be with a different microphone.

Since this is an XLR microphone I also had to pick up an audio interface. For this I went with the Focusrite Scarlett Solo 4th generation. I thought about a few others and even the previous generation but ended up going this way in part to have the XLR connection in the back and also because this is a trusted piece of hardware. Just for fun I ended up getting a green XLR cable. This serves no practical purpose but I think it’s fun.

Setting it up is easy, I had to take down the Yeti and plug a few things in. Originally I left the shock mount I was using for the Yeti on but after using it a while I don’t think it was necessary and the shock mount is kinda big so I’m seeing if I notice a difference for a while.

I did encounter some issues using it at first as the recordings were only appearing as a right channel instead of both left and right. I did some digging and found that the Focusrite software doesn’t have a linux version. However, there is a ALSA Scarlett GUI created by Geoffrey Bennet that allows for some software based adjustments. When I got this installed I needed a firmware update that was easy enough to install. With this I was able to access the Scarlett Solo and direct audio from the SM57 to both left and right channels. It would be weird if I was only ever talking in everyone’s right ear.

With that figured out I tried a few recordings just to make sure things were working correctly and experimenting with my technique. One thing I know I needed to do was reevaluate my audio chain and adjust it for the properties of this new microphone. I’m currently using OBS to apply audio filters on my microphone before it ever goes out into Teams, Zoom, or whatever else I might use a microphone for. It’s mostly used to talk with Matthew through Overwatch for example. Anyway, this chain is as follows:

• Gain - I try to keep the gain on the microphone or audio interface fairly low and boost the gain digitally. My understanding is that this is to prevent clipping at the start.
• Noise Suppression - RNNoise in OBS does a great job in my experience.
• EQ (cuts) - Here is where problematic areas in the sound are removed. I can sound nasaly at times so I typically have some cuts in the mid range to try and address this issue.
• Compressor - Prevents peaking if I get too loud like when I yell about what our team is doing when playing Overwatch with Matthew as mentioned above.
• Limiter - Currently set to ensure sounds never go above -2 dB.
• EQ (flavor) - Adds something back in to make me easier to understand. I find I need more presence so there’s a bit of a boost in both the high and low ends.

I ended up boosting the Gain a bit more in OBS with this microphone as even with better technique audio was coming in a bit quiet. I spent a bit more time fine tuning exactly where in the frequency certain aspects were and reducing them. This was all new since previously I was only using the 3-Band EQ that comes with OBS. I managed to install some VSTs that included a parametric equalizer which offers way more control. Below is an image of the cuts that I’m currently doing for an example.

I’ll probably end up doing some tests again in a few days just to see if everything sounds okay still. I don’t think I sound the same every time I try this so I’m hoping for an average that I’m happy with.

One last thing, I glossed over the fact that I’m using OBS in the middle. This is achieved by running a .sh file every time my computer starts up. The file creates a virtual speaker, a virtual mic, and links them together.

# Create a virtual sink that can be set as a monitor in OBS
pactl load-module module-null-sink sink_name="VirtualSpeaker" sink_properties=device.description=VirtualSpeaker

# Link it with a virtual source that is visible in pulseaudio apps like Zoom
pactl load-module module-null-sink media.class=Audio/Source/Virtual sink_name="VirtualMic" channel_map=front-left,front-right

#Link together
pw-link VirtualSpeaker:monitor_FL VirtualMic:input_FL
pw-link VirtualSpeaker:monitor_FR VirtualMic:input_FR

Once this is run I can select VirtualSpeaker as the Monitor device within OBS and as the recording device in conferencing software or games. This is specifically a Linux based solution to this issue. If you’re on Windows or Mac you can just install virtual cables and have a similar experience. I ran this setup on both Windows and Mac prior to moving to Linux.

I find the audio experience is often poor on conference calls and I don’t want to contribute to that. This flow isolates my voice so people just hear me and not my environment. I have a mechanical keyboard that can be quite loud at times, but so long as I’m not talking at the same time as I’m typing, you would never know that I’m typing with an unmuted microphone. I like to mute myself anyway but I wouldn’t have to which is nice.

This is day 56 of #100DaysToOffload.