Local voice dictation for macOS.
Nothing leaves your machine.

Download

[ demo video — TBC ]

What it does

Hold & speak

Hold Right Option, speak, release. Text appears at your cursor.

On-device STT

Runs Parakeet or Whisper speech-to-text entirely on your Mac.

Local LLM correction

Fixes grammar, punctuation, and technical terms. No API keys needed.

Code-aware

Detects your editor, language, and file. Applies correct naming conventions.

Dev terms built in

80+ developer term corrections out of the box. Extensible with your own rules.

Voice assistant

Ask questions with Claude, Codex, Ollama, or a local LLM. Speaks answers back.

Purpose-built for vibing

Dictate code, prose, messages, docs — whatever you're working on. The pipeline completes in under 2 seconds.

What it doesn't do

Send your audio or text anywhere.
Require an account, login, or API key.
Need an internet connection.

Before / After

Raw STT output vs. what Super VoiceMode produces.

Raw so um import the num pie library and create a data frame
Output import the numpy library and create a DataFrame
Raw add a function called validate token
Output add a function called validate_token
Raw so um we need to do three things first update the database second run migrations and third restart the server
Output We need to do three things:
1. Update the database
2. Run migrations
3. Restart the server

Personas

What started as a hidden easter egg became too much fun to keep secret.

Super VoiceMode ships with 30+ personas for the voice assistant. Each one changes how the assistant thinks and talks — its personality, vocabulary, and style. You can stack up to three at once, and every combination produces wildly different results.

💅 Valley Girl + 🚀 Tech Bro
"okay so like, we literally need to disrupt this codebase and it's giving scalable"
🦴 Caveman + 💼 Corporate
"fire good. circle back on fire next quarter"
🫖 Butler + 💀 Gen Z
"Sir, that PR is no cap bussin. Shall I merge at your convenience?"

Some are genuinely useful — Scientist keeps answers precise, Corporate nails the professional tone. Most are just hilarious. All of them run locally.

Some personas are hidden. If you know, you know.

Characters — coming soon

Personas change how the assistant talks. Characters go further — they have backstories, memory, opinions, and a voice all their own. Create your own or use community characters. More details soon.

How it works

  Audio Capture
       ↓
  Voice Activity Detection (Silero VAD)
       ↓
  Speech-to-Text (Parakeet / Whisper — on-device)
       ↓
  Deterministic Replacements (fillers, tech terms, numbers)
       ↓
  LLM Correction (Qwen 1.7B — on-device, optional)
       ↓
  Clipboard → Paste

Every step runs locally. The entire pipeline completes in under 2 seconds.

Pricing

Currently in beta — free while we build

Standard

$XX / year

Voice dictation, on-device STT, LLM correction, voice assistant, built-in personas.

14-day free trial. No credit card required.

Pro

$XX / year

Everything in Standard, plus custom voices, custom characters, advanced personas, and all future Pro updates.

14-day free trial. No credit card required.

Download

Download for macOS

macOS 14+ · Apple Silicon