SrizonVoice
A lightweight macOS menu bar dictation app that converts speech to text. Powered by Groq Whisper AI with LLM post-processing (Groq or Gemini) and support for 111 languages — hold a hotkey, speak, release to transcribe.
Requires macOS Ventura (13.0) or later • Free to use with your own API key
“Apple could not verify” Warning
Since SrizonVoice is not notarized by Apple yet, macOS Gatekeeper may block the app on first launch. To fix this, open Terminal and run:
xattr -cr /Applications/SrizonVoice.appThis removes the quarantine flag that macOS adds to downloaded apps. You only need to do this once.
See It in Action
Clean, native macOS interface that stays out of your way
Live Demo — Hold hotkey, speak, release to transcribe
Menu Bar Controls
API Key & Permissions
Transcription Settings
AI Post-Processing
Powerful Features
Fast AI transcription in a simple menu bar app
AI Post-Processing
Transcripts are refined by an LLM to fix grammar, capitalize URLs, and clean up speech quirks. Choose from GPT-OSS 120B, GPT-OSS 20B, Llama 3.3, or Google Gemini.
Hold-to-Dictate
Hold the hotkey while speaking, release to transcribe. Audio is sent to Groq Whisper for fast, accurate speech-to-text.
Global Hotkey
Default Fn key works from any app — fully customizable in Settings. Press Escape while recording to cancel without transcribing.
111 Languages
Dictate in any of 111 supported languages with country flag selectors. Recently used languages appear in the menu bar for quick switching.
Works Everywhere
Text is inserted directly into any active text field using Accessibility APIs, with enhanced support for Electron apps like Slack, Discord, Notion, and more.
Lightweight & Native
Built with Swift and SwiftUI as a native menu bar app. Minimal resource usage, launches at login, and stays out of your way.
Bring Your Own Key
Use your own Groq and optional Gemini API keys stored securely in macOS Keychain. Audio is sent directly to providers — no middleman servers.
How It Works
Three simple steps to hands-free typing
Hold Hotkey
Hold the Fn key (or your custom shortcut) from any app to start recording
Speak Naturally
Talk at your natural pace — a floating island shows a live waveform while you speak
Release to Transcribe
Release the hotkey and text is transcribed, refined by AI post-processing, and inserted into your active text field
Quick Setup
Get started in under a minute
Download the DMG file and open it
Drag SrizonVoice to your Applications folder
Launch the app — it appears in your menu bar
Enter your Groq API key (and optionally a Gemini API key for post-processing), then click "Grant Permissions" to set up Microphone & Accessibility access
Hold your hotkey (default: Fn) and start dictating!
Note: You'll need to grant Microphone and Accessibility permissions in System Settings for dictation and text insertion to work.
111 Languages Supported
Dictate in virtually any language with country flag selectors and quick-switch your recent languages from the menu bar
🇺🇸
English
🇪🇸
Spanish
🇫🇷
French
🇩🇪
German
🇨🇳
Chinese
🇯🇵
Japanese
🇸🇦
Arabic
🇧🇩
Bengali
🇮🇳
Hindi
🇧🇷
Portuguese
🇷🇺
Russian
🇰🇷
Korean
🇮🇹
Italian
🇹🇷
Turkish
🇻🇳
Vietnamese
🇳🇱
Dutch
🇵🇱
Polish
🇹🇭
Thai
🇸🇪
Swedish
🇮🇩
Indonesian
And 91 more languages supported
Requirements
System
- macOS Ventura (13.0) or later
- Microphone permission
- Accessibility permission
API Keys
- Groq API key (required) — free tier available
- Gemini API key (optional) — for Google Gemini post-processing
Privacy & Security
SrizonVoice respects your privacy. Audio is only captured when you hold the hotkey and is sent directly to Groq servers for transcription. Post-processing text is sent to Groq or Google depending on your choice. No data passes through Srizon servers.
See Groq Privacy Policy, Groq Terms of Use, and Google Privacy Policy for AI provider policies.
Ready to Dictate?
Download SrizonVoice and start typing with your voice across any macOS app today.
Download SrizonVoice