Voice input for Cursor, Claude Code and VS Code
Prompts are getting longer than the code. FnFlow lets you say them instead. Hold fn in the Cursor chat, the Claude Code terminal or any editor, talk for as long as you need, then let go.
How it works
- Click into Cursor's composer, Claude Code, VS Code, Windsurf or any other text field.
- Hold fn and describe what you want: the context, the constraints, the bug you saw. For a long prompt, double-tap fn to go hands-free.
- Let go. The punctuated text appears in about 1.2 seconds and FnFlow checks that it landed. If it didn't, it's one click away in history.
Why it fits coding
- On a test sentence, adding five terms to the vocabulary turned “update Zapot” into “update Saypad” and “post-GIRS” into “Postgres”.
- Cursor and VS Code are Electron editors that hide their text fields from macOS until something asks. FnFlow asks before it pastes.
- Whisper large-v3-turbo runs on your GPU, so nothing about your codebase leaves the Mac.
- There's no weekly word quota for long prompts to burn through.
Prompts that are easier to say than type
- “Refactor the auth middleware so tokens refresh before expiry, keep the public API the same, and add tests for the 401 path.”
- “Explain why this query is slow and suggest an index – the table has about forty million rows.”
- “Write a migration that splits the name column into first and last name, and backfill existing rows.”
Questions
What if voice input in Cursor isn't working?
FnFlow doesn't depend on Cursor's own voice input. It records and transcribes on your Mac, then pastes into the field you're in, so it works in the composer and the terminal panel either way. If nothing appears, check that FnFlow has Microphone and Accessibility permissions, and look in its history, where every dictation is saved.
Does it work in the terminal?
Yes. FnFlow pastes with ⌘V like you would, so a terminal gets the text the same way any other app does. We test mostly in Cursor and the Claude app. If a paste ever goes missing, the text is one click away in history.
How do I get project names right?
Add them in Settings → Vocabulary, for example “Supabase, Postgres, Zustand”. Whisper uses them as spelling hints on every dictation.
Will it send my prompts anywhere?
No. Recognition runs on your Mac. Only what you paste into Cursor or Claude goes where those apps send it.
Local Wispr Flow alternativeFnFlow vs VoiceInkHow 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 developersFnFlow vs EnviousWisprHow 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