{"version":"2.1.59","anchor":"voice-dictation-mode","canonical_anchor":"voice-dictation-mode","heading":"Voice Dictation Mode","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.59\/e\/voice-dictation-mode","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.59","markdown":"### Voice Dictation Mode\n\n**What**\n\nHold the Space bar to dictate text via speech-to-text, with real-time transcription streamed over WebSocket to Claude.ai's `\/api\/ws\/speech_to_text\/voice_stream` endpoint.\n\n**Usage**\n\n`\/voice # Toggle voice mode on\/off` Then hold Space to record; release to stop. The transcript is injected into the input field.\n\n**Details**\n\n- Requires a Claude.ai account (OAuth); run `\/login` first if not signed in\n- Persisted via the `voiceEnabled` setting in `settings.json`\n- Audio capture uses a bundled native module (`audio-capture.node`) on macOS\/Linux, with fallback to `sox` (`rec` command) or `arecord` on Linux\n- Offers guided SoX installation (detects `brew`, `apt-get`, `dnf`, `pacman`) if no audio tool is found\n- Supports a \"focus mode\" that continuously records when the terminal has focus, flushing each finalized segment immediately\n- Cursor displays an animated audio-level indicator (bar graph characters ` \u2581\u2582\u2583\u2584\u2585\u2586\u2587\u2588`) with color cycling while recording\n- Shows \"listening\u2026\" while recording and an animated \"Voice: processing\u2026\" while finalizing\n- Supports 11 languages: English, Spanish, French, Japanese, German, Portuguese, Italian, Korean, Chinese, Hindi, Indonesian \u2014 configured via the `language` setting\n- Microphone permission checks with platform-specific guidance (macOS: System Settings \u2192 Privacy & Security \u2192 Microphone; Linux: system audio settings; Windows: Settings \u2192 Privacy \u2192 Microphone)\n- Not available in remote\/headless environments \u2014 prints helpful message directing users to run locally\n\n**Evidence**\n\n`\/voice` command (search for `\"Toggle voice mode\"`), WebSocket client (search for `\"\/api\/ws\/speech_to_text\/voice_stream\"`), settings definition (search for `\"voiceEnabled\"`)\n\n**[Gradual Rollout]** \u2014 Gated behind the `tengu_amber_quartz` feature flag (defaults to `!1`\/false). The `\/voice` command is hidden when the flag is off. Users with the flag enabled who are logged in to Claude.ai can use this feature."}