OpenWhispr
Share
What is OpenWhispr?
OpenWhispr is an open-source voice dictation app that turns your speech into text inside any program. Press a hotkey, talk, and the words appear wherever your cursor is: an email, a document, Slack, or a ChatGPT, Claude or Cursor prompt. It runs on macOS, Windows and Linux, with an iPhone and iPad app on paid plans.
Local or cloud transcription
Its standout trait is that it works offline. You download a speech recognition model (several OpenAI Whisper sizes, from about 75 MB up to 1.6 GB, or NVIDIA Parakeet) and the audio is processed on your own machine without going to any server. When speed matters more, you can switch to the OpenWhispr Cloud service or plug in your own OpenAI, Anthropic, Gemini or Groq keys. It recognizes more than 100 languages and detects which one you're speaking.
Beyond dictation
- AI cleanup and voice commands, such as turning rough notes into a polished email.
- A personal dictionary that picks up names and jargon from your corrections.
- Meeting notes without adding a bot to the call, with speaker labels.
- Audio file transcription, plus chat over your notes on the Business plan.
Privacy and licensing
The source code is on GitHub under the MIT license, so anyone can check how data is handled. The company behind it, Gizmo Labs Inc., says cloud audio is deleted as soon as transcription finishes and that your dictation is never used for training. It also claims SOC 2 Type II, ISO 27001, HIPAA and GDPR compliance, which makes it relevant for people dictating sensitive material, from clinicians and lawyers to teams with strict security policies.
It positions itself as an open alternative to paid dictation apps like Wispr Flow or SuperWhisper, with one key difference: local dictation is unlimited and free for good.
OpenWhispr features and use cases
How to use OpenWhispr
OpenWhispr FAQ
Is OpenWhispr really free?
Yes. Dictation with local models is unlimited and free, and so is using your own API keys. You only pay for unlimited transcription on their cloud or other Pro and Business features.
Does it work offline?
Yes. With a local model, all transcription happens on your computer and no connection is needed.
Is my voice sent to a server?
Not with a local model. If you use the cloud, the company says audio is deleted once transcription finishes and isn't used to train models.
Which apps can I dictate into?
Anything that accepts text, including Word, Google Docs, Gmail, Slack, Notion, your terminal and the ChatGPT or Claude chat box.
Does it support languages other than English?
Yes. It handles more than 100 languages and can auto-detect the one you're speaking.
Which speech model should I pick?
It depends on your hardware. Smaller Whisper models are faster and lighter; larger ones such as Whisper Turbo are more accurate but take up about 1.6 GB.
Can it take meeting notes?
Yes. It records and transcribes meetings without joining a bot to the call, and labels speakers. The Free plan includes 5 hours a month.
Is there a mobile app?
There's an iPhone and iPad app included with the Pro and Business plans.
Is it really open source?
Yes. The code is on GitHub under the MIT license, which allows personal and commercial use.
Are there any discounts?
Students, teachers, military, non-profits and people who rely on it for accessibility can get 40% off for 12 months.
OpenWhispr pricing and plans
Reference price: 8 USD / month. Source · Verified: October 19, 2026
Prices in US dollars (USD), checked on the official website on 19/10/2026.
Free: $0
- Unlimited local dictation and unlimited transcription with your own API keys.
- 2,000 OpenWhispr Cloud words a week and 5 hours of meeting recordings a month.
- Custom dictionary and community support. No card needed.
Pro: $8/month or $80/year
- Unlimited cloud transcription and agent mode.
- 20 hours of meetings a month, cross-device sync, personal API, MCP integration and the iPhone and iPad app.
- Email support.
Business: $16/user/month or $160/user/year
- Everything in Pro, plus unlimited meetings, chat over your data and priority support.
Enterprise: custom pricing with SSO, SAML, SCIM, audit logs, retention controls and data processing agreements.
Students, military, non-profits and people using it as an assistive tool can request 40% off for 12 months by emailing support. Plans can be changed or cancelled at any time.
OpenWhispr Screenshots
OpenWhispr alternatives
AI ReFounder
24/7 AI voice agent for online stores. Automatically handle calls, resolve inquiries, and close sales with seamless Shopify integration.
Linden by Speechmatics
Low-latency speech-to-text for AI agents. Detects turns, identifies speakers, and captures names and numbers with high precision in over 55 languages.
ThunderPhone
High-precision voice AI for call automation. Supports 40+ languages, handles interruptions, and offers SIP integration for customer service and sales.
SpeechTurbo
Convert audio and video to text with high accuracy and speed. Supports 98 languages, AI summaries, and translation. Pay-as-you-go with no monthly subscriptions.
