This guide assumes you already have VoiceMeeter setup and if you haven’t please refer to my previous post about setting up Voicemeeter with OBS. Also to note this explains some of the VST settings however, if you are looking for an in-depth explanation of each plugin than you’ll have to check that our elsewhere or maybe in a future guide.
You’ll need the below 2 pieces of software:
- Cantabile Lite
- Reaper VST’s
Configuring Cantabile Lite
First lets start with installing Cantabile lite, navigate to
https://www.cantabilesoftware.com/download/ and install the Cantabile software, this installer includes different versions but we’re going to make use of the Cantabile Lite (Free) version.
Lets first start by heading into Tools > Options > Audio Engine.
We’re going to first start by selecting “VoiceMeeter Potato Insert Virtual ASIO”.
As for buffer size, i’m using 512 as that is what my Microphone buffer is defaulted to, (also my only option). If you have more than one buffer rate here I would suggest starting with something in the middle and later on adjusting as you see fit. As for “Double Buffered Audio”, I have this enabled as it reduces the chances of CPU spikes affecting the audio at the cost of an increase of buffer time. Again this is something you will have to decide whether to enable or not.
Next head into the Audio ports tab and we’re going to set up our Microphone in’s and Out’s
Select Add and choose Stereo input port I’ve named mine “Mic IN” this will create a Mic input with two un-assigned channels, select the left channel and assign it to the left channel, then do the same for the right.
Now do the same but instead use “Stereo Output” and assign it in the same way, once completed you should have something similar to the below.
Download the Reaper VST plugins and install them, once installed make a note of where they have been installed to.
In my case this was C:/Program Files/VST Plugins/ReaPlugs
Lets set the plugin location in the “Plugin Options” tab
Patching in VoiceMeeter
Lets now patch this into Voicemeeter, to do this is pretty simple. In VoiceMeeter’s
As our microphone is
Let’s start by adding our first plugin, we’re going to use the Routing view to do this as it’s easier to visualise the route than the conventional list way. In the top right you can switch between the two views.
Click on the big plus and select add plugin, we’re going to choose ReaFir as our first plugin to eliminate background noise, this will be noises such as computer fans and other ambient noise that you might not notice.
Once selected we’re going to click on “Mic IN” and drag it onto “Stereo In”, as you can see the output is being set automatically to “Mic Return”
Lets now set the Mic in to send audio to Stereo in on the ReaFir plugin.
Lets now configure ReaFir, double click on the plugin to open its configuration options
Set the mode to “Subtract” and then enable “Automatically build noise profile” for about 30 seconds, make sure to be silent during this to allow the plugin to correctly pick up the background noise as we don’t want it to filter out other audio.
Next, we’re going to set up ReGate, this is a noise gate that will remove any audio below a certain threshold such as mouse and keyboard clicks. Import the plugin and route it as shown below.
Double click the plugin to bring up the settings for it.
The first thing we want to do is use the slider on the left to select the threshold for when audio is cut out, so adjust the slider so that keyboard and mouse clicks cannot be heard and your voice can. This will take a little adjusting to get right.
The next thing we’re going to change is the Release time to 200 ms, this is purely a personal preference so that it doesn’t cut the Microphone out as soon as you’ve finished speaking.
I’ve also changed the Highpass to 90 Hz, again this is a personal preference as if you have a lower voice you may find that without it the filter affects the sound of your voice.
The Output mix
We’re using two of the bands in this case so let’s go again and disable 3 and 4. For our first band we’re going to put a High Shelf pass on it with a Frequency setting of between 2000-2500 Hz, then use a bandwidth of around 2-3 Octaves.
For band 2 we’re using a High pass filter to come some of the “boom” from our voice using a frequency of around 60 Hz. As well as using a wide bandwidth of 3 Octaves.
These settings will be completely depending on the type of voice you have as well as the microphone and accessories you are using such as a pop filter.
Finally lets add our ReComp VST into the chain, we’re using this so that it levels the volume that your voice comes out at, for example, if you’re playing a horror game talking normally but then shout in fear due to a jump scare the audio will automatically be lowered so that your microphone doesn’t peak.
We’re going to first start by settings the Attack time to 5 ms and the release to 50 ms as I feel it’s a good balance between them.
I’m using a ratio of 5:1, which means that for every 5dB over the threshold then the compressor will lower it to 1dB above the threshold. For example see the below table.
Using the bar on the left we’re going to adjust the threshold, whilst speaking in a low voice we want the red bar on the right to be between -3 and -6, and then speaking louder adjust the audio so that the compression is between -6 and -12.
Finally, we’re going to use the wet makeup on the right to adjust the gain increase we’re going to add to compensate for the compression of our voice, in my case I don’t need it, however, you may wish to adjust this higher or lower depending on the other tweaks you’ve made.
Now you’ve added this you can turn OFF the audio engine and compare the difference from before and after,
You may notice sometimes when you start up Cantabile that your audio is robotic or sounds broken, this is a compatibility issue some people have with using VoiceMeeter and Cantabile at the same time and can be remedied by simply turning the Cantabile audio engine off and on a few times.
Additionally for the sake of stability it is a wise idea to start Cantabile Lite first before then launching Voicemeeter.