Local and offline transcription
Use local speech-to-text models on the computer for offline dictation. OpenWhispr lists Whisper models ranging from Tiny to Turbo, with different size and accuracy profiles.
OpenWhispr
OpenWhispr is open-source dictation and speech-to-text software for macOS, Windows, Linux, and iOS. It helps developers, writers, teams, and people seeking hands-free input dictate into existing apps, transcribe audio, and create meeting notes using local models or optional cloud processing.
Open-source dictation and transcription
OpenWhispr is open-source dictation and speech-to-text software for turning spoken input into editable text. It supports local AI models for on-device transcription, offline use, and privacy-sensitive workflows, while optional cloud transcription provides another processing path.
The desktop app can send dictated text to the active cursor in writing, communication, and development applications. It also supports audio uploads, meeting recording and notes, voice-driven cleanup or drafting, and multilingual transcription. The product is available across macOS, Windows, Linux, and iOS, with additional sync and team capabilities on paid plans.
Core capabilities
Dictation across applications
Dictate into the application already in use, including editors, email, documents, chat tools, and development environments. Text is returned at the active cursor for review.
Voice instructions and AI cleanup
Turn spoken instructions into formatted output, such as cleaning up an email or drafting a message, using the product’s voice-driven agent mode.
Meeting notes and audio transcription
Transcribe recordings and capture meeting notes without requiring a meeting bot to join the call. Meeting outputs can include transcripts, notes, decisions, action items, and open questions.
Multilingual speech-to-text
Support 100+ languages with automatic language detection, including conversations that switch languages.
Custom vocabulary
Add specialized names, jargon, and technical terms through a custom dictionary. The site also describes learning from corrections to improve recognition of recurring words.
Practical workflows
Developers writing in an IDE
Speak code comments, documentation, commit messages, or prompts while staying in an IDE such as Cursor. The resulting text can be reviewed at the active cursor instead of being typed manually.
Teams capturing meeting notes
Record or transcribe a meeting and review structured notes covering decisions, action items, and unresolved questions. Local processing can be used when a meeting bot or cloud upload is not appropriate.
Writers and professionals drafting text
Draft articles, newsletters, emails, briefs, or other long-form text by speaking naturally, then revise the returned text in the writing application. A custom dictionary can help with recurring names and terminology.
Hands-free and accessible computer use
Use on-device dictation to reduce keyboard use or make speech input available in the apps already being used. The use-cases site specifically describes workflows for accessibility, RSI, arthritis, dyslexia, and mobility-related needs.
Transcribing recordings and research
Upload audio or use local transcription to turn lectures, interviews, podcast episodes, or other recordings into searchable text, then dictate summaries or follow-up copy into another application.
Pros and Cons
Pros
- Unlimited local dictation is available on the free plan.
- Local processing supports offline use and keeps audio on the user’s computer when local models are selected.
- Dictation works at the active cursor across a broad range of writing, chat, email, and developer applications.
- The product combines speech-to-text with meeting notes, audio uploads, custom vocabulary, and voice-directed text cleanup.
- OpenWhispr is open source under the MIT license and supports 100+ languages.
Cons
- The free OpenWhispr Cloud allowance is limited to 2,000 transcribed words per week, although local dictation is unlimited.
- Cross-device sync, the mobile companion app, API and MCP access, and agent mode are paid-plan features rather than Free-plan features.
- Local transcription requires choosing and running a model on the user’s device, so the experience may vary with the selected model and computer.
FAQ
Which platforms does OpenWhispr support?
OpenWhispr runs on macOS, Windows, Linux, and iOS. The Free plan starts on desktop, while the iPhone and iPad companion apps are included with Pro and higher plans.
What is the difference between local and cloud transcription?
Local transcription runs on the computer using local AI models, so audio does not need to be sent to OpenWhispr. Cloud transcription is an optional path; the site also says recordings are deleted after transcription and that transcription data is not used to train models.
Can I use my own API keys?
Yes. OpenWhispr supports local AI models and lets users bring their own API keys for cloud transcription. The free plan includes unlimited local dictation and cloud transcription with your own keys.
Where does the dictated text go?
OpenWhispr places dictated text at the active cursor in compatible applications, including tools such as ChatGPT, Claude, Cursor, Google Docs, Gmail, Slack, and Microsoft Teams. Users can then review and edit the text in that application.
What does the free plan include?
The free plan includes unlimited local dictation, unlimited local AI models, 100+ languages, a custom dictionary, and five hours of meeting recordings per month. It also includes up to 2,000 words per week of OpenWhispr Cloud transcription; paid plans add higher cloud and meeting limits plus features such as sync and team tools.
Quick Facts
- Category
- Dictation and speech-to-text software
- Platforms
- macOS, Windows, Linux, and iOS
- Processing
- Local AI models, OpenWhispr Cloud, or supported user API keys
- Languages
- 100+ languages with automatic detection
- Source model
- Open source under the MIT license
- Primary workflows
- App-wide dictation, meeting notes, audio transcription, and voice drafting
Альтернативы OpenWhispr
SpeakoFlow
SpeakoFlow is a free, open-source desktop voice-to-text app for Windows, macOS, and Linux. It supports dictation into any app, voice-driven writing with Flow, cleanup, translation, and a screen-aware assistant.
QuickQuill
QuickQuill is a macOS dictation and transcription app that runs locally on the device. It helps users record meetings, transcribe audio, generate summaries, and export notes without using a cloud service.
Zen Whisper
Zen Whisper is a local-first dictation and transcription app for Apple Silicon Macs running macOS 14 or later. It lets users dictate into most Mac text fields, transcribe media or supported public links, and work across 110 spoken language options.
Speech to Text Converter
Speech to Text Converter is a browser-based transcription tool for live dictation and uploaded audio or video files. It offers a free tier for short tasks and a Pro plan for unlimited transcription, AI summaries, translation, speaker identification, and advanced exports.
CAKIE
CAKIE is a Mac app for recording your screen and voice for an AI agent. It adds a transcript and screenshot highlights, then lets you copy a prompt and paste or attach the context in another AI app.
Sanota
Sanota is an app that turns spoken memories, reflections, and interviews into clear written stories. It supports personal storytelling, family history, and shared memories, with guided prompts and subscription pricing.