Your voice never reaches us
When you dictate locally, your words are transcribed on your own computer by open models. Nothing is sent to us or to anyone else.
Your voice is 3x faster than your keyboard.
Open source voice-to-text assistant. Private by design. Write faster in every app.
Loading...Available on macOS, Windows and LinuxWithout OpenWhispr
Slack — #product
Type a message...
With OpenWhispr~150 WPM
Slack — #product
Type a message...
Sprint Planning
Recording
SC
Sarah ChenJL
James LeeMK
Maria KimAC
YouSprint Planning
TranscriptNotesEnhanced
Mar 11, 2026
00:00
AI Chat
Hold ⌥Space to speak
Open source & privacy first
Your data stays yours.
OpenWhispr is open source. Here’s how we handle your data:
0%
Data Retention
Cloud is a choice, not a default
Switch the cloud on and your audio goes out to be transcribed, then it's gone the moment the words come back. We never keep the recording.
Data Training
Off by default
We don't train on what you say
Your transcriptions are not used to train models or improve the product. If that ever changes, it will be opt-in and we will ask first.
Don't take our word for it
OpenWhispr is open source on GitHub under the MIT license. Every claim on this page is one you can go and check for yourself.
Compliance & certifications
Everything your security team will ask for, already in place.
Compliant
HIPAA
Whether you use local or cloud processing, we keep your health information safe.
SOC 2 TYPE II
ISO 27001
Compliant
AICPA SOC 2 & ISO 27001
Independently verified. Whatever your security review needs to see, we already have it.
What you get
Dictation, AI cleanup, offline mode, and 100+ languages.
Your machine. Your models.
Run local speech-to-text models or bring your own API keys. All free.
Whisper Tiny
Fastest75 MB
Whisper Base
Fast142 MB
Whisper Small
Balanced466 MB
Whisper Medium
Accurate1.5 GB
Whisper Turbo
Best1.6 GB
Give instructions by voice.
Say things like 'clean this up' or 'draft an email to Mike' and it does exactly that.
You said
"hey OpenWhispr, clean up this email — I need to tell mike that the meeting is moved to thursday at 3 and also remind him about the deck"
Output
Hi Mike,
Quick heads up — our meeting has been moved to Thursday at 3 PM. Also, just a reminder to have the deck ready.
Thanks!
It learns your words.
Names, jargon, and terms the AI might miss — auto-learned from your corrections.
Add a word or phrase...
+
OpenWhisprDr. Smith×gRPC×Kubernetes×Synty×PostgreSQL×
Auto-learns from your corrections
Speak any language.
100+ languages with auto-detection. Switch mid-conversation, we keep up.
🇺🇸English
🇪🇸Español
🇫🇷Français
🇩🇪Deutsch
🇯🇵日本語
🇨🇳中文
🇧🇷Português
🇷🇺Русский
🇰🇷한국어
🇮🇹Italiano
🇸🇦العربية
🇮🇳हिन्दी
🇹🇷Türkçe
🇳🇱Nederlands
🇵🇱Polski
🌐100+ more
Works across all your apps.
ChatGPT, Claude, Cursor, Slack, Docs — if it accepts text, it works.
Works offline.
No wifi, no problem. OpenWhispr runs offline so you can transcribe anytime, anywhere.
Common questions
What people usually want to know before they download.
What people are saying
Real feedback from developers, writers, and teams using OpenWhispr.
My favorite thing on @producthunt today is OpenWhispr. I love @aquavoice_ but really need a local model for when I'm traveling or don't have WiFi. This perfectly solves it. UI and Performance are great.
OpenWhispr is goated
Translated from Japanese
The best voice input solution I've found recently. I was using Aqua Voice before, but this is OSS with the same features and feel — plus you can choose between API-based pay-as-you-go or downloading a local ASR model!
Reading is faster than listening, and speaking is faster than writing. Makes me think the highest throughput person-to-AI mode will be voice to text and reading / evaluating outputs. I'm trying this with OpenWhispr for voice to text.
OpenWhispr is a really good open source alternative for WisprFlow!
Whisper Base local is near instant in OpenWhispr, and I found it better than Parakeet. Parakeet is near instant, too.
I think it's changed the game when you can talk to it and then it can talk back. I'm using OpenWhispr (found on GitHub) with OpenAI Whisper Turbo and it works seamlessly.
Started using OpenWhispr with my local models & hardware, I have to say this is better than typing.
Translated from Japanese
I've gotten tired of the hassle of typing into Claude Code chats, so I tried using OpenWhispr for voice input instead. Even when I stumble and rephrase mid-sentence, the transcription comes out clean and polished, which makes me happy.