Voice input in Google Antigravity

Published 6 November 2026

Antigravity has two voice features. Dictation types what you say into the message box, and you start it with the microphone or Ctrl+M. Voice Mode is a spoken conversation with the agent. The CLI has both as well, under /voice. Here's how each works and what to check when it fails, taken from Google's docs. We didn't test Antigravity ourselves.

The controls

Antigravity AppAntigravity CLI
DictationMicrophone in the message box, or Ctrl+MF5, /voice dictate or /record
Voice ModeStart Voice mode button, shown when the message box is emptyF6 or /voice live
Stop dictatingClick the microphone or press Ctrl+M againEnter or F5 keeps the text, Esc discards it

From antigravity.google/docs/features, /docs/voice-mode and /docs/cli/commands/voice, checked 6 November 2026.

When to use which

Use dictation for a prompt you want to read before you send it. That's most prompts with file names in them. In the CLI, if the box already has text, the transcript gets appended. Ctrl+Z afterwards undoes the whole dictation in one step.

Use Voice Mode when you want to work hands-free: give a task, steer while the agent works and approve permission prompts out loud. Voice Mode continues the conversation you're in, so the agent keeps the context of what you typed earlier.

If voice input isn't working

  1. Microphone permission. On a Mac, open System Settings → Privacy & Security → Microphone and make sure Antigravity, or the terminal running the CLI, is on.
  2. Wrong input device. Check the input device in System Settings → Sound. A disconnected headset is a common cause.
  3. CLI mode. Dictation needs an interactive session and is disabled in --print mode.
  4. Account type. Google's docs say CLI dictation isn't currently supported for business or enterprise accounts.
  5. Stuck on Connecting. The status line should move from Connecting… to Recording. If it doesn't, check your network, since the transcription depends on a service.
  6. Over SSH. A remote machine has no access to your microphone. Google's mic-serve subcommand streams it through a reverse SSH tunnel, and the CLI must be installed on both ends.

When you want it local, or everywhere

Antigravity's voice lives inside Antigravity and depends on a service. FnFlow is a free Mac app that runs Whisper on your machine: hold fn, say the prompt, release, and the text is pasted into Antigravity's message box or anything else. It works without internet and won't talk back. Dictating prompts for AI coding tools has tips for any of these.

Dictate prompts locally, in any tool.Free · macOS 15 or later · Apple Silicon · 187 MB
Download FnFlow

Questions

How do I dictate in Antigravity?

In the app, click the microphone in the message box or press Ctrl+M. In the CLI, press F5 or type /voice dictate or /record. The text stays in the box until you send it.

What's the difference between dictation and Voice Mode?

Dictation gives you text to review and edit, and the agent answers in writing. Voice Mode runs on Gemini Live and the agent answers out loud while its work appears in the conversation.

Why is Antigravity voice input not working?

Check that the app has microphone access, and in the CLI that you're in an interactive session and not on a business or enterprise account, where dictation isn't currently supported. Over SSH, the remote machine can't reach your microphone without mic-serve.

Does Antigravity transcribe locally?

Google describes it as using the latest Gemini Audio models, and the CLI docs describe a transcription service. We found no offline option in the docs, so assume your audio leaves your Mac.

More from the blog
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 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 inputVibe coding with voice input