FnFlow vs EnviousWispr
EnviousWispr is the closest app to FnFlow we've compared. It's free, needs no account and recognizes speech on your Mac. It's also open source under GPLv3 and adds an optional AI polish step. FnFlow leaves that step out and focuses on getting text into Cursor and terminals reliably. EnviousWispr's details come from its site and GitHub README, checked 10 October 2026.
At a glance
| FnFlow | EnviousWispr | |
|---|---|---|
| Price | Free | Free; cloud polish uses your own API key |
| Open source | No | Yes, GPLv3 (the EG-1 model has its own license) |
| Where speech is processed | On your Mac (whisper.cpp, Metal) | On your Mac (Parakeet v3 or WhisperKit) |
| Model download | Small model inside the 187 MB installer; 874 MB turbo model in the background | Parakeet during setup; Whisper turbo about 1.64 GB on demand |
| AI rewriting | None | Optional: on-device (EG-1, S1-mini, Apple Intelligence, Ollama) or your own cloud key |
| Hotkey | Hold fn or right ⌘, double-tap for hands-free | Keybind you set; push-to-talk, toggle or double-press hands-free |
| History | Every dictation with its audio | Dictations and file transcripts, searchable |
| Account | Not needed | Not needed |
| macOS | 15 or later, Apple Silicon | 14 or later, Apple Silicon |
EnviousWispr details from enviouswispr.com and its GitHub README, checked 10 October 2026.
Where EnviousWispr is better
- You can read the code, build it and change it. There's also a Homebrew cask:
brew install --cask saurabhav88/tap/enviouswispr. - Parakeet gives you a fast engine for European languages, and you can switch to WhisperKit when you need wider language coverage.
- If you do want cleanup, the polish step can stay on your Mac. Its EG-1 model is pitched at turning spoken lists into real lists and keeping only the corrected half of a self-correction. If polish fails, you get the raw text back.
- It transcribes audio and video files and can split them into speaker turns.
- Silero VAD ends a recording when you stop talking, and a cancelled dictation's text is kept for 24 hours.
Where FnFlow is different
The first dictation works offline straight after install. The Whisper model is already inside the 187 MB download, and the more accurate turbo model arrives in the background on Wi-Fi.
There is no rewriting stage to configure or to second-guess. A prompt for Claude Code goes in as you spoke it, and the project names you added to Vocabulary reach Whisper as spelling hints (up to 60 of them).
- FnFlow checks the paste. In Cursor, VS Code and Slack it first sets Electron's
AXManualAccessibility, then reads the field back and tells you whether the text landed. - History keeps the audio of every take, so you can replay it or, on macOS 26, re-run it with Apple's SpeechAnalyzer engine.
- Push-to-talk lives on fn, or right ⌘ when macOS already uses the 🌐 key, and ordinary shortcuts like ⌘C don't trigger a recording.
Switching
Give the push-to-talk key to one app at a time. Copy your custom words from EnviousWispr's dictionary into FnFlow's Vocabulary by hand, since there's no import. If you like AI polish for prose, keep EnviousWispr for that and use FnFlow for prompts and commit messages.
Questions
Is EnviousWispr really free?
Yes. Its site says it's free and open source, with no account and no subscription. Cloud polish runs on your own API key, so the provider may charge you, with no EnviousWispr markup.
Which speech models do they use?
EnviousWispr offers Parakeet v3, downloaded during setup and aimed at 25 European languages, and WhisperKit running Whisper Large v3 Turbo, about 1.64 GB on demand with 99+ languages. FnFlow uses whisper.cpp with large-v3-turbo in an 874 MB 8-bit build, plus a small model inside the installer.
Do either of them collect analytics?
Both send anonymous data you can switch off. EnviousWispr's README says PostHog analytics and Sentry crash reports are on by default and can be turned off in App Settings → Privacy. FnFlow sends anonymous usage counts with no ID and no text, and they can be turned off too.
Are you neutral?
No, we make FnFlow. We only used EnviousWispr's own site and repository.
Local Wispr Flow alternativeFnFlow vs VoiceInkVoice input for CursorHow 7 Mac dictation apps get you to your first dictationWhisper.cpp on Apple Silicon: which large-v3-turbo quantization to shipHow I built a local dictation appVoice dictation in Claude Code and the terminalWhy holding fn doesn't work for dictation on Mac (and how to fix it)How to run Whisper locally on an Apple Silicon MacSpokenly alternative for developersHow to dictate a good prompt for AI coding toolsVoice dictation in VS Code on a MacTeaching Whisper your jargon: initial prompt and replacementsWhy Whisper types "Thank you" on silence, and how to filter itPrivate dictation on a Mac at work: what actually leaves your machineVoice input for Codex CLI and the ChatGPT appVoice coding on a Mac: what actually worksWriting git commit messages by voiceDictating code review comments on GitHubDictation in IntelliJ IDEA, PyCharm and WebStorm on a MacDictation in Zed on a MacDictation in Terminal, iTerm2 and WarpWhich Whisper model to use on an Apple Silicon MacPush-to-talk dictation on a Mac: hold, toggle or streamSuperwhisper alternative for developersClaude Code voice mode: how to enable /voice and fix itWhisper apps: five ways to run Whisper locallyTalon voice coding vs a dictation appPush to talk on a MacDoes Claude have a voice mode?VoiceInk vs Wispr FlowVoiceInk vs SuperwhisperParakeet vs Whisper on a MacApple SpeechAnalyzer vs WhisperWhisper.cpp vs faster-whisper on a MacMlx-whisper vs whisper.cppGitHub Copilot voice inputGemini CLI voice inputVoice input in Google AntigravityVibe coding with voice input