BridgeVoice
Desktop Dictation.
100% On-Device.
Runs Whisper AI locally to transcribe your speech instantly into any app. Write code, docs, and Slack messages natively with your voice. Zero cloud dependency, total privacy.
See Voice Coding in Action
Watch how our builders use BridgeVoice to dictate code, comments, and Slack messages with sub-second latency directly into any application.
The Engine for Voice Coding
Built with Rust and Whisper AI. Every feature designed for builders who ship fast and demand privacy.
Universal Text Injection
Transcribed text is instantly pasted into whatever app has focus — your code editor, terminal, Slack, Notion, email, or browser. Works everywhere, not just inside one tool.
100% On-Device Processing
Powered by Whisper AI running locally in Rust. Zero cloud dependency, zero data collection, zero telemetry. Your voice never leaves your machine — ever.
Push-to-Talk & Toggle
Two recording modes with fully customizable global hotkeys. Hold to dictate or toggle on/off — your workflow, your way.
Sub-Second Transcription
Persistent audio stream starts recording in under 10ms. End-to-end from speech to text in under 1 second.
Custom Developer Dictionary
Teach BridgeVoice your specific terminology. Define replacements for technical terms, product names, or variable conventions that standard Whisper models might get wrong.
Usage Stats & History
Every transcription is saved to local history. Track total words, speaking time, and words-per-minute. One click to recopy anything you dictated.
Speak to inject text into any application.
BridgeVoice doesn't force you into a specific editor. By acting as a global text injection layer, it works beautifully with VS Code, Cursor, your terminal, Slack, and your browser.
- Dictate Commit Messages & CommentsYou are deep in a coding session. Instead of breaking flow to type a commit message or inline comment, hold your hotkey, speak naturally, and the text appears exactly where your cursor is.
- Write Docs, Emails & Slack MessagesEvery builder writes more than code — documentation, pull request descriptions, Slack replies, emails. Dictate all of it at the speed of thought without leaving the app you are in.
- Accessible DevelopmentRepetitive strain injuries, carpal tunnel, and other conditions make keyboard-heavy work painful. BridgeVoice provides an immediate, ergonomic alternative for text input across your entire desktop.
- Offline & On-The-GoOn a plane, in a coffee shop with unreliable Wi-Fi, or behind a firewall — BridgeVoice works exactly the same. Full on-device Whisper transcription with zero internet required.
Built for Privacy & Speed
Most dictation tools send your voice to the cloud, introducing latency and compromising your privacy. BridgeVoice wraps the hyper-optimized whisper.cpp engine in a native Tauri application to give you immediate, private transcription.
- Native OS integration via Tauri 2.0
- Hardware-accelerated inference where available
- Works completely offline without Wi-Fi
- Cross-platform: macOS, Windows, Linux
Frequently Asked Questions
What is BridgeVoice?
BridgeVoice is a privacy-first desktop voice dictation app built with Tauri 2.0 and Rust. It runs OpenAI's Whisper model locally on your machine to transcribe speech and injects the text directly into whatever app you are using — your code editor, terminal, chat, browser, or anything else.
Is my voice data actually private?
Yes — this is not a marketing claim. BridgeVoice processes all audio entirely on your device using whisper.cpp compiled in Rust. No audio samples are transmitted over the network. No telemetry. No analytics on your speech. You can use it in airplane mode with full functionality.
How do I get access to BridgeVoice?
BridgeVoice is included with the BridgeMind Pro subscription ($20/month or $16/month billed annually). Pro unlocks BridgeVoice plus BridgeMCP, BridgeCode, BridgeSpace, Premium Skills, and the full Prompt Library.
What platforms does it run on?
BridgeVoice runs natively on macOS, Windows, and Linux. It is built with Tauri 2.0 and Rust for native performance. macOS builds are code-signed and notarized by BridgeMind LLC.
Which Whisper models are available?
BridgeVoice supports 6 Whisper model sizes: Tiny (75 MB, fastest), Base (142 MB), Small (466 MB, recommended), Medium (1.5 GB), Large-v3 (3.1 GB, highest accuracy), and Turbo (fast large model). Choose based on your accuracy needs and hardware.
Do I need special hardware?
BridgeVoice works with any microphone, including built-in laptop mics. It handles stereo-to-mono conversion and sample rate resampling automatically. For best results, a quality headset or external mic is recommended.
Ready to Ship with Your Voice?
BridgeVoice is included with BridgeMind Pro — along with BridgeMCP, BridgeCode, BridgeSpace, and the full toolkit. One subscription, every product.
AVAILABLE FOR MACOS, WINDOWS, AND LINUX