Desktop Dictation · 100% On-Device

Voice to Text at the
Speed of Thought

Runs Whisper AI locally to transcribe your speech instantly into any app. Press a hotkey, speak, and your words appear at your cursor — cleaned up and ready to go.

Get Started Free → See How It Works

Sub-second latency

Vulkan GPU acceleration

100% offline capable

Built with Rust + Tauri

Core Features

The Engine for Voice Dictation

A complete voice-to-text toolkit that works offline, respects your privacy, and makes your words smarter.

Universal Text Injection

Press a hotkey, speak, and text appears at your cursor in any application. No window switching, no copy-pasting.

100% On-Device Processing

All transcription happens locally. Your audio never leaves your machine. Works completely offline.

GPU Accelerated

Vulkan-powered Whisper engine. Sub-second transcription on NVIDIA, AMD, and Intel GPUs with CPU fallback.

AI Processing Modes

Clean up grammar, draft emails, summarize, or translate — all from your voice. Powered by GPT-4o and Claude.

Configurable Hotkeys

Push-to-talk with fully customizable global hotkeys. Default Alt+Space — change it to whatever works for you.

Usage Stats & History

Track your dictation history, review past transcriptions, and see statistics on your usage patterns over time.

How It Works

Three steps. That's it.

Press Your Hotkey

Hit Alt+Space (or your custom hotkey) from anywhere. HyperVoice activates instantly in the background.

Speak Naturally

Talk normally. Whisper AI transcribes your speech locally in real-time with GPU acceleration.

Text Appears at Cursor

Your transcribed text is automatically pasted into whatever app you were using. Optionally AI-processed first.

Pricing

Simple, transparent pricing

Local transcription is free forever. Subscribe for AI-powered cloud processing modes.

Free

$0 /forever

Everything you need for local dictation.

Unlimited local transcription
GPU-accelerated Whisper (Vulkan)
All 6 Whisper models
Configurable hotkeys
History & statistics

Download Free

Popular

HyperVoice Pro

$8.99 /month

Everything in Free, plus AI cloud processing.

Everything in Free
AI cloud processing (clean up, email, summarize, translate)
GPT-4o & Claude-powered modes
Bring your own API keys (optional)
Priority support

Subscribe Now

FAQ

Frequently asked questions

Does HyperVoice send my audio to the cloud?

No. All speech-to-text transcription happens 100% locally on your machine using Whisper AI. Your audio never leaves your device. Cloud processing is only used for optional AI post-processing modes (like grammar cleanup or translation), and you can disable those entirely.

What GPUs are supported?

HyperVoice uses Vulkan for GPU acceleration, which works with NVIDIA, AMD, and Intel GPUs. If no compatible GPU is found, it automatically falls back to CPU-based transcription. GPU mode dramatically reduces transcription time.

What Whisper models are available?

HyperVoice supports all standard Whisper model sizes: Tiny (75 MB), Base (142 MB), Small (466 MB), Medium (1.5 GB), Large-v2 (2.9 GB), and Large-v3 (3.1 GB). Larger models are more accurate but require more VRAM and processing time.

Can I use HyperVoice offline?

Yes! Core voice-to-text transcription works completely offline once you've downloaded a Whisper model. You only need an internet connection for the optional AI cloud processing modes (clean up, email, summarize, translate).

What platforms are supported?

HyperVoice is currently available for Windows. Built with Tauri 2.0 and Rust, it's designed as a native desktop application for maximum performance and minimal resource usage. macOS and Linux support is planned.

Do I need a subscription to use HyperVoice?

No. Local voice-to-text transcription is completely free with no limits. The HyperVoice Pro subscription ($8.99/mo) adds AI-powered cloud processing modes that use GPT-4o and Claude to clean up, format, summarize, or translate your transcriptions.

Voice to Text at the Speed of Thought