Voice to Text, Privately.

On-device speech transcription for macOS, Windows & Linux

Press a hotkey, speak, and your words appear as text. All processing happens locally on your device using whisper.cpp. No cloud. No accounts. No tracking.

Download for macOSDMG Download for Windowsx64 · Installer Download for Linuxx64 · AppImage

Free to start · Also available as .deb for Debian/Ubuntu · All downloads on GitHub

Everything you need to turn voice into text

Wspr lives in your system tray and works with any app. Powered by whisper.cpp with GPU acceleration for fast, accurate, on-device transcription.

Live Transcription

Press a global hotkey from any app. Speak naturally and your words are transcribed in real time, then inserted at your cursor or copied to clipboard.

Hotkey & Push-to-Talk

Ctrl+Shift+Space by default, fully customizable. Tap to toggle recording, or hold for push-to-talk — release and your transcription appears. Separate hotkey for transcribe + rewrite.

AI Text Rewriting

Polish transcriptions with AI. Built-in styles for minimal, professional, or concise output. Use OpenAI, Anthropic, or Google Gemini with your own API key.

File Transcription

Transcribe audio or video files alongside live dictation — file and voice pipelines run concurrently, so a 2-hour meeting transcript won't block your hotkey. MP3, WAV, M4A, MP4, MOV, MKV and more.

99 Languages

Whisper models support 99 languages with automatic detection. Transcribe English, French, Spanish, Japanese, Arabic, and many more.

Cross-Platform

Runs on macOS, Windows, and Linux. Same features, same privacy, everywhere you work.

Diagram of Wspr's on-device pipeline: audio captured by the microphone stays inside your device, with no cloud upload, no accounts, and zero tracking.

Your data stays on your device

Wspr processes everything locally. Your voice never leaves your computer. No servers, no accounts, no analytics, no tracking.

  • On-device transcription via whisper.cpp with GPU acceleration
  • Audio never leaves your computer
  • API keys stored securely in system keychain
  • Zero third-party analytics or ad SDKs
Wspr AI rewrite: a raw transcript with filler words on the left is transformed into a polished, professional sentence on the right, with style chips for Professional, Minimal, Concise, and Custom, powered by your OpenAI, Anthropic, or Gemini API key.

Polish your words with AI

Automatically clean up transcriptions with built-in rewrite styles, or create your own. Choose your preferred AI provider.

  • Minimal, Professional, and Concise built-in styles
  • OpenAI, Anthropic, and Google Gemini
  • Toggle rewrite on/off from the recording overlay
  • Custom rewrite styles with Pro
Whisper model picker showing speed and accuracy for Tiny (75 MB), Base (140 MB), Small (460 MB, recommended), Medium (1.5 GB), and Large-v3 (3 GB).

Choose speed or accuracy

Wspr includes five OpenAI Whisper models. Pick the one that fits your workflow — from lightning-fast to maximum accuracy.

  • Tiny (~75 MB) — Fastest, basic accuracy
  • Base (~140 MB) — Fast, good accuracy
  • Small (~460 MB) — Balanced, recommended
  • Medium (~1.5 GB) — High accuracy
  • Large-v3 (~3 GB) — Best accuracy

One price. Yours forever.

No subscription. No recurring fees. Pay once and own Wspr Pro for life.

Free

$0

Get started instantly

  • 50 live transcriptions
  • 5 file transcriptions
  • Built-in rewrite styles
  • Cloud AI rewrite (own API key)
Recommended

Pro

$14.99

One-time purchase. No subscription.

  • Unlimited live transcriptions
  • Unlimited file transcriptions
  • Custom rewrite styles
  • Unlimited history

The best value in voice-to-text

Most voice-to-text apps charge monthly subscriptions, skip Linux entirely, or cost 2–5x more up front. Wspr gives you everything — on-device, cross-platform, AI rewriting — for a single $14.99 payment, forever.

Feature Wspr Wispr Flow Superwhisper MacWhisper
Price $14.99 lifetime $12–$15/mo Subscription / Lifetime $29–$80
Platforms macOS, Win, Linux macOS, Win, iOS, Android macOS, Win, iOS macOS
On-device transcription Cloud
AI rewrite with your own key OpenAI, Anthropic, Gemini Proprietary Multiple (BYO key) OpenAI / Claude
Live / system-wide dictation
File transcription Not advertised
One-time purchase available ✓ (Lifetime tier)

Common questions about Wspr

Is Wspr really free to start?

Yes. Wspr includes 50 free live transcriptions, 5 file transcriptions, and built-in AI rewrite styles at no cost. No credit card or account required. Upgrade to Pro for $14.99 (one-time) to unlock unlimited everything.

Does my voice data stay on my device?

100%. All speech-to-text transcription is processed locally on your device using whisper.cpp with GPU acceleration. Your audio never leaves your computer. Cloud AI is only used for optional text rewriting — and only if you choose to enable it with your own API key.

What platforms does Wspr support?

Wspr runs on macOS, Windows (x64), and Linux (x64). All features are available on every platform.

Why pay $14.99 when competitors charge $10–15 per month?

Because one month of Wispr Flow costs the same as a lifetime of Wspr. Pay once, own it forever — no recurring charges. You get on-device transcription, AI rewriting, and unlimited usage for a single payment.

How is Wspr different from built-in dictation?

Built-in dictation has limited accuracy, no custom vocabulary, and can't rewrite your text with AI. Wspr uses OpenAI's Whisper models for superior accuracy across 99 languages, supports custom vocabulary hints, and offers AI-powered text rewriting with OpenAI, Anthropic, or Google Gemini.

Does Wspr support push-to-talk?

Yes. Toggle push-to-talk in Settings → General. With it enabled, hold your transcribe hotkey to record and release to transcribe — radio-style. Quick taps still work as toggles, and the same flow works on macOS, Windows, and Linux.

Try Wspr for free.

Download now and get 50 free transcriptions. Upgrade to Pro for $14.99 — one-time, forever.

Download Wspr Buy Pro — $14.99