HyperWhisper transcribes audio from your microphone. It does not directly capture system audio (the audio coming out of your speakers — Zoom calls, YouTube videos, podcasts, etc.). If you want to transcribe something playing on your computer, you have two options.Documentation Index
Fetch the complete documentation index at: https://hyperwhisper.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
Option 1 — Use a virtual audio cable (recommended)
A virtual audio cable is a small driver that creates a “fake” microphone that captures your system audio. Once installed, you can pick it as HyperWhisper’s input device and transcribe anything playing on your Mac or PC.- macOS
- Windows
The most popular free option is BlackHole (open source, free):Paid alternatives like Loopback offer a friendlier UI and per-app routing.
Install BlackHole 2ch
Download from existential.audio/blackhole and run the installer.
Create a Multi-Output Device
Open Audio MIDI Setup (Spotlight → “Audio MIDI Setup”). Click + → Create Multi-Output Device, then tick both your Built-in Output (or speakers/headphones) and BlackHole 2ch.This lets you hear the audio while BlackHole captures it.
Set the Multi-Output Device as your system output
In System Settings → Sound → Output, pick the new multi-output device.
Pick BlackHole as HyperWhisper's input
In HyperWhisper, open Settings → Audio Input and select BlackHole 2ch.
Option 2 — Transcribe a recording after the fact
If you can record the meeting or video to a file (e.g. Zoom’s local recording, OBS, QuickTime, screen recorder), you can drop the file into HyperWhisper directly:- macOS: Click the menu bar icon → Transcribe File → pick your audio or video file.
- Windows: Right-click the tray icon → Transcribe Audio File → pick your
.wav,.mp3, or.m4a.
Native system-audio capture (without a virtual cable) is on our roadmap. If this is important to you, let us know at support@hyperwhisper.com — it helps us prioritize.
