> ## Documentation Index
> Fetch the complete documentation index at: https://superwhisper.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# "No voice found" or stuck processing

> Fix dictations that show "No voice found" or never finish processing. Check your microphone, input level, and connection.

**"No voice found"** means the recording had no detectable speech in it. History shows **No voice found in recording** for these dictations. The problem is almost always the audio input, not the model.

## Fix "No voice found"

<Steps>
  <Step title="Check which microphone is active">
    The active input device shows in the **top-right of every settings pane**. If Superwhisper is listening to the wrong device, such as a disconnected headset or a muted external microphone, switch it there or set **Use system default**.

    <img style={{ borderRadius: '0.5rem' }} src="https://mintcdn.com/superwhisper/do-XBuvOUgvVmZf6/images/screens/mac/input-device-dropdown.png?fit=max&auto=format&n=do-XBuvOUgvVmZf6&q=85&s=31978731705ddafb96422f794ae2715a" alt="Superwhisper input device selector open from the top-right of the settings window" width="1760" height="2060" data-path="images/screens/mac/input-device-dropdown.png" />
  </Step>

  <Step title="Confirm microphone permission">
    On macOS, open **System Settings → Privacy & Security → Microphone** and make sure Superwhisper is enabled. On Windows, check **Settings → Privacy & security → Microphone**.
  </Step>

  <Step title="Watch the waveform while you speak">
    Start a recording and talk. If the waveform doesn't move, audio isn't reaching the app. Try the built-in microphone to rule out a Bluetooth device problem.

    One exception: when a mode records **system audio**, the waveform stays still by design. Check [History](/docs/get-started/history) to confirm audio was captured.
  </Step>

  <Step title="Play back the recording">
    Open the dictation in **History** and play the audio. If the recording is silent, the input device was wrong or muted. If you can hear yourself clearly but still got the error, [report the recording from History](/docs/get-started/history#report-an-issue). It sends the audio to the team with your report.
  </Step>
</Steps>

<Note>
  Check that **Automatically increase microphone volume** is on in **Settings → Sound**. It sets the microphone input volume to maximum when a recording starts, so quiet input is less likely to read as silence. It works only when Superwhisper uses the system default device.
</Note>

## Stuck on processing

If dictations hang at the processing stage instead:

1. **Check your connection**: Cloud models need internet. Try a local model to confirm.
2. **Check the model finished downloading**: A local model that's still downloading can't transcribe. Open **Settings → Models library**.
3. **Restart Superwhisper**: retry the dictation from [History](/docs/get-started/history). Your audio is already saved.
4. **Report it**: turn on **Error logging** in **Settings → Configuration**, reproduce the problem once, then [report the recording from History](/docs/get-started/history#report-an-issue).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.