Editorial image illustrating Isolating System Audio: How to Record Echo-Free Zoom Sessions with Virtual Drivers
Every actor has been there. You finish a grueling 90-minute coaching session via Zoom, packed with adjustments you need to memorize for tomorrow's callback. You hit "Record" in your capture software, hoping to review the director's notes later. When you play it back, the audio is a disaster: your voice sounds distant, the coach’s voice echoes, and a hollow reverb makes the recording unusable for any serious review.
The issue is rarely the microphone itself. It is the routing. When you record a streaming platform like Zoom directly through your system's default settings, your computer creates a feedback loop. It hears the audio coming out of the speakers and immediately feeds it back into the microphone input, creating that tinny, cavernous echo. Buying a more expensive USB microphone will not fix this structural problem. To fix it, you need to separate the audio playback from the recording chain using a virtual audio interface.
I have tested dozens of workflows for the Castandapps lab, and the most reliable method for capturing clean system audio—specifically the other person's voice on a call—involves installing a virtual driver. This acts as a digital patch bay, routing audio internally without it ever hitting the air in your room.
Here is the precise process to set this up on both macOS and Windows.
Why External Hardware Isn't Always the Fix
Many actors assume the solution is an audio interface like a Focusrite Scarlett or a mixer like the Rodecaster Pro. While these are excellent for capturing high-fidelity voice, they do not solve the software routing issue if you are trying to record a remote session. Plugging a mixer into your computer still relies on your operating system's sound management, which defaults to that dreaded loopback configuration.
Furthermore, we are discussing practice tapes and coaching reviews here. While I advocate for high-end production on final submissions, a review file is a utility. You need to hear the nuances in the coaching, not necessarily hear your own voice in Grammy-winning quality. The goal is intelligibility and separation. We want the Zoom audio (the coach) to be recorded on a separate track or at least cleanly isolated from your room noise.
Selecting the Right Virtual Driver for Your OS
The software landscape changed significantly in 2024, and as of 2026, two tools dominate the market for stability. For macOS users, the standard is BlackHole. For Windows users, Voicemeeter Banana remains the industry standard for its versatility.
These tools create "virtual" inputs and outputs. Your computer thinks they are real headphones or microphones, but they exist entirely in code. This allows you to send audio from Zoom to the virtual device, and then tell your recording software to listen to that same device, bypassing your speakers entirely.
Configuring BlackHole on macOS
macOS users have an advantage here because BlackHole is lightweight and integrates natively with the Core Audio architecture. It requires no system restarts once installed.
- Download and Install: Navigate to the Existential Audio website and download the latest version of BlackHole. You will likely want the 2-channel version for standard self-tapes, though a 16-channel version exists if you are doing complex multitracking. Install the
.pkg file.
- Open Audio MIDI Setup: This is the hidden control center for Mac audio. Press Command + Space and type "Audio MIDI Setup" to find it.
- Create a Multi-Output Device: Click the tiny + symbol in the bottom left corner of the window and select Create Multi-Output Device.
- Configure the Routing: You will see a list of your audio devices in the sidebar. Check the boxes next to Built-in Output (your headphones/speakers) and BlackHole 2ch. This tells the Mac to send audio to both places at once: to your ears so you can hear, and to the BlackHole driver so the software can record it.
- Set System Output: Right-click (or Control-click) on your new Multi-Output Device in the sidebar and select Use This Device For Sound Output.
Your audio is now split. You can hear the call, but the system is also sending a clean copy to the BlackHole driver.
Setting Up Voicemeeter Banana for Windows
Windows requires a slightly more robust interface because its audio handling is less centralized. Voicemeeter acts as a mixer board.
- Install Voicemeeter Banana: Download the standard installation package from VB-Audio. You will need to restart your computer to install the virtual drivers properly.
- Set Default Playback: Right-click the speaker icon in your Windows taskbar and go to Sound Settings. Set your default output to Voicemeeter Input (VB-Audio Voicemeeter VAIO). This sends all computer sound into the mixer.
- Configure Hardware Out: Open the Voicemeeter application. In the top right section, labeled Hardware Out, click A1 and select your headphones (WDM or MME driver). This ensures the audio comes out of your headphones so you can hear it.
- Enable the Virtual Input: Look at the middle section of the mixer, B1. You should see the green level meters moving when you play a system sound. This indicates audio is flowing into the software.

Bridging Zoom into Your Recording Software
With the drivers running, we just need to tell Zoom and your recording software (QuickTime, OBS, or Audacity) where to send and receive the audio.
For Zoom:
- Open Zoom Settings > Audio.
- Set Speaker to your Multi-Output Device (Mac) or Voicemeeter Input (Windows).
- Set Microphone to your actual physical mic (if you want to record your voice too) or leave it as is if you only care about the incoming audio.
For your Recording Software (using OBS Studio as the example):
- Open Audio Mixer > Settings.
- Add an Audio Input Capture. Select the virtual driver: BlackHole 2ch on Mac or Voicemeeter Output (VB-Audio VoiceMeeter AUX) on Windows.
Now, when you record in OBS, you are capturing exactly what Zoom is playing, directly from the digital stream. No room echo, no feedback.
The Trade-off: Latency and Monitoring
There is one caveat to this setup you must be aware of: latency. Because the audio is being processed through a virtual driver before it hits your "real" headphones, there is a tiny delay—usually 10 to 20 milliseconds.
In a coaching session, this is rarely noticeable. However, if you are trying to sing along to a track or read in perfect sync with a pre-recorded voice, you might notice the timing is slightly off. You can adjust this in the Audio MIDI Setup on Mac by changing the buffer size to 256 samples for a lower delay, though this puts a slightly heavier load on your CPU.
If you are using this method for practice tapes, ensure your visual frame still adheres to broadcast standards. Just because the audio is a "hack" doesn't mean the lighting should be amateur. As we covered in our analysis of ring lights and key lighting setups, the visual quality still dictates the casting director's initial impression, even for practice takes.
Finalizing the File for Delivery
Once the session is recorded, you will have a clean video file with isolated audio. If you used OBS, you can export this immediately.
If you are submitting this practice tape to a coach for review or uploading it to a cloud service, be mindful of file naming and format. I see too many actors lose opportunities simply because their submission links are broken or their file formats are non-standard. When you send that file, ensure you avoid the common pitfalls that lead to rejection in automated submission forms.
The Strategic Advantage of Clean Audio
Why go through this trouble for a practice tape? Because professional habits are formed in the rehearsal room, not just on the set. By training yourself to demand clean audio isolation, you develop an ear for what the casting director hears. You start to notice the hum of the refrigerator or the echo of the tile floor that you previously ignored.
Setting up a virtual driver removes the technological barrier between you and the performance. It allows you to focus entirely on the acting rather than fighting with your equipment. When you can listen back to a coaching session and clearly hear the inflection in your own voice without the distraction of digital distortion, you improve faster. In 2026, the actors who treat their home studios like broadcast booths—even for practice—are the ones who book the rooms.