superwhisper not working on a Mac: what each symptom means, and the fix

By Ryleigh Newman · Published October 6, 2026

When superwhisper stops working on a Mac, the cause is usually one of five things: Accessibility permission, which it needs to paste into other apps; the wrong microphone, which shows up as No voice found; a shortcut another app already owns; a model that is still downloading or needs a connection; or a Mac still on macOS 13, since current releases need macOS 14. Your words are usually safe while you fix it: superwhisper says it saves every dictation as you speak, so you can reprocess it from History afterwards.

I built YapToText, a different dictation app for the Mac, and it comes up once, near the end. Everything about superwhisper here comes from its own troubleshooting docs, read on October 6, 2026, and each page is linked under Sources. I have not reproduced these problems myself, so the steps are superwhisper's, in my words.

This app is completely free. You don't have to make an account. There are no subscriptions, no difficult downloads, and nothing is locked. Made for Apple silicon Macs. Source on GitHub.

This guide in other languages: العربية, Čeština, Dansk, Deutsch, Ελληνικά, Español, Suomi, Français, עברית, हिन्दी, Magyar, Bahasa Indonesia, Italiano, 日本語, 한국어, Norsk, Nederlands, Polski, Português, Română, Русский, Slovenčina, Svenska, ไทย, Türkçe, Українська, Tiếng Việt, 简体中文, 繁體中文

On this page: 10 sections

1. Steps
2. Match what you see to a fix
3. If it says No voice found, or gets stuck processing
4. If nothing pastes, or it pastes the wrong text
5. If the shortcut does nothing
6. If it will not open, or crashes
7. If it is slow, or types the wrong language
8. Where YapToText fits
9. The short version
10. Questions

Steps

1. Check the active microphone and Microphone permission if it says No voice found.
2. Turn on Accessibility if nothing pastes.
3. Reset the shortcut in Settings, Configuration if it does nothing.
4. Reprocess the dictation from History once it is fixed.

Match what you see to a fix

From superwhisper's troubleshooting docs, read October 6, 2026. Turning on Error logging in Settings, Configuration before you reproduce a problem makes any report useful.

Fix the cause first, then reprocess the dictation from History instead of saying it again.

If it says No voice found, or gets stuck processing

superwhisper says No voice found means the recording had no detectable speech, and that the problem is almost always the input, not the model. Find out whether any sound reached the app before you change models.

1. Check the active microphone. It shows at the top right of every settings pane. Click it to switch, or choose Use system default.
2. Check the permission. System Settings, Privacy & Security, Microphone, with superwhisper on.
3. Watch the waveform. Start a recording and talk. If it does not move, no audio is arriving. Try the built-in microphone to rule out a Bluetooth device. A mode that records system audio keeps the waveform still on purpose.
4. Play the recording back. Open it in History. Silence means the wrong or muted input. Clear speech with the error is worth reporting from History.
5. Let it raise the volume. Automatically increase microphone volume, in Settings, Sound, sets the input to maximum when a recording starts, when superwhisper uses the system default device.

If dictations hang on processing instead, superwhisper's list is: check the connection, because cloud models need one; check that a local model finished downloading, in Settings, Models library; restart and retry from History, where the audio is already saved.

If nothing pastes, or it pastes the wrong text

When a dictation lands in History but not in your app, the transcription is fine and only the delivery failed. Accessibility is the most common cause, according to superwhisper.

1. Grant Accessibility. System Settings, Privacy & Security, Accessibility, with superwhisper on. It needs this to paste into other apps.
2. Keep the cursor where it was. superwhisper pastes at the active cursor, so clicking away while it processes leaves the paste nowhere. The text is still on the clipboard.
3. Apps that block pasting. Secure fields, some terminals and virtual machines refuse synthetic pastes. Simulate keypresses, in advanced settings, types the result key by key. superwhisper calls it experimental and US QWERTY only.
4. Old clipboard text. With Clipboard content set to Keep what I have copied, the default, a slow app can read your old clipboard back. Choose Replace with dictation instead.
5. Pasted twice. Usually Simulate keypresses plus an app that also accepts the paste, or the shortcut pressed twice while processing.

If the shortcut does nothing

superwhisper has two kinds of shortcuts, and mixing them up is the cause it sees most. Global shortcuts live in Settings, Configuration, and each mode can have its own.

1. Another app owns the key. Global shortcuts lose to system and app shortcuts. Try an uncommon combination to confirm, then pick a key nothing else uses.
2. A single modifier key. Command or Fn can work alone, but superwhisper notes some keyboards send Fn to the system instead of apps. Test a regular combination.
3. The recording window is set to None. Recording still works but shows nothing on screen. Dictate a line and look in History.
4. Reset it. Clear the shortcut and record it again in Settings, Configuration, then quit and reopen superwhisper.
5. Look for a clash. System Settings, Keyboard, Keyboard Shortcuts, for the same combination.

If a key starts the wrong mode, the same key is probably set both as a global shortcut and as a mode's own shortcut. Give each its own key.

If it will not open, or crashes

Two different problems look alike here: the app refusing to open, and the app quitting mid-dictation. Check the macOS version first: current superwhisper releases need macOS 14 or later.

1. On macOS 13. superwhisper's docs say version 2.16.1 is the last release for macOS 13.3, listed with older releases at superwhisper.com/appcast.xml. Its download page still said 13.3 on October 6, 2026.
2. Reinstall over the top. Download it again from superwhisper.com and drag it into Applications. Your settings live in the Superwhisper folder, not the app, so replacing the app is safe.
3. Restart the Mac and try again. If it crashes at launch, superwhisper asks for an email with your macOS version and the crash report.

For crashes during or after a dictation, superwhisper says the audio is written to disk as you record, so the dictation usually waits in History. Its steps: update from Settings, Configuration; if a local model is in use, switch the mode to a cloud model or a smaller one such as Whisper Base, because a crash that stops means the Mac ran short of memory; test the plain Voice to Text mode; and free up disk space.

If it is slow, or types the wrong language

superwhisper names four things behind slow dictations: your Mac's resources, the connection, the model, and whether AI processing runs. The model you pick, and how long it stays loaded, usually matter most.

• Voice model active duration. In Settings, Advanced settings, under Voice model, from 10 seconds to 1 hour. Shorter frees memory but reloads the model each time.
• A smaller or cloud model. Whisper Base, or a cloud model such as S1-Voice, takes the load off the Mac.
• Skip the AI step. The Voice to Text mode transcribes without AI processing.
• English instead of your language. With the language on Automatic, superwhisper says non-English speech sometimes comes out in English. Set the mode's language to yours, or add its suggested hint phrase in your language to Vocabulary.

If you get raw, unformatted text where a mode should have formatted it, the AI step failed. superwhisper notes most language models run in the cloud even when the voice model is local, so reconnect or switch the mode's language model to a local one such as S1-mini, then reprocess from History.

Where YapToText fits

superwhisper and YapToText are close cousins: both run Whisper on your Mac. The difference that matters here is that YapToText's cleanup also runs on the Mac, free, with nothing to download first.

Both models ship inside YapToText's App Store app, so there is no model download to wait for and no cloud language model to fail offline. It is Mac only, needs Apple silicon and macOS 14, and has no Windows or phone app. The superwhisper alternative page sets out what superwhisper does that it does not.

The short version

• No voice found. Check the input device and Microphone permission, then play the recording back in History.
• Nothing pastes. Turn on Accessibility, and keep the cursor where it was.
• Shortcut dead. Another app owns the key. Reset it in Settings, Configuration.
• Will not open. Current releases need macOS 14. 2.16.1 is the last for macOS 13.3.
• Crashes. Update, then try a cloud or smaller model.
• Raw text. The AI step failed: reconnect or pick a local language model, then reprocess.

Fix it, then reprocess from History. superwhisper keeps the audio. If something here no longer matches superwhisper's docs, tell me on GitHub.

This app is completely free. You don't have to make an account. There are no subscriptions, no difficult downloads, and nothing is locked. Made for Apple silicon Macs. Source on GitHub.

About the author. Ryleigh Newman is a developer whose hands make typing difficult and who dictates every day with YapToText, the app this site describes. Facts about other apps come from their own sites, with the date they were read. More about the project.

Related

• superwhisper keeps crashing or will not open: how to find the cause, and where your dictation went
• superwhisper pasted raw text: the AI step failed, and here is how to get the formatted version
• superwhisper guides: every page on this site about superwhisper, by topic
• superwhisper is slow or uses too much memory: the settings that change it
• superwhisper uses the wrong microphone, or system audio records silent: the fixes
• superwhisper Realtime is grayed out or missing: the three reasons, and how to turn it on
• superwhisper keyboard shortcuts: global keys, per-mode keys, push to talk, and deep links
• superwhisper adds words, repeats text or answers a question: find the stage, then fix it
• Change the language in superwhisper, and stop it writing English when you spoke something else
• superwhisper says No voice found, or gets stuck processing: how to fix it

Sources

• https://superwhisper.com/docs/common-issues/troubleshooting (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/no-voice-found (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/llm-processing-error (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/app-wont-launch (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/crashes (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/pasting-issues (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/shortcuts-not-working (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/microphone-and-system-audio (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/performance-tips (checked 2026-10-06)
• https://superwhisper.com/docs/common-issues/language-detection (checked 2026-10-06)
• https://superwhisper.com/download (checked 2026-10-06)
• https://github.com/ryleighnewman/YapToText (checked 2026-10-06)
• https://apps.apple.com/us/app/yaptotext/id6786382289 (checked 2026-10-06)

Questions