OpenTypeless
Free AI voice input toolSpeak, and AI writes it in 99 languages
Highlights
What you can do with it
- Dictate and polish emails directly in Gmail.
- Capture quick voice notes in Slack with cleaner wording.
- Write code comments using a custom technical dictionary.
How to use it
- 1 Press the global hotkey and pick a language.
- 2 Speak naturally while the provider transcribes.
- 3 Let the AI polish grammar and format.
- 4 The finished text appears at your cursor.
Who it is for
Pros
- Completely free and open source.
- Many providers and 99 languages to choose from.
- Context-aware output fits the app you are in.
Limitations
- Open-source and free terms could change; check the repo.
- Setup means choosing providers and configuring hotkeys.
- Accuracy varies with microphone quality and provider choice.
What OpenTypeless is
OpenTypeless is a free, open-source desktop app for AI voice input. You press a global hotkey, speak, and the app transcribes your words, polishes the text, and inserts it at your cursor. It supports up to 99 languages with auto-detection, and it works with several speech-to-text providers such as Deepgram Nova 3, AssemblyAI, and GLM ASR. Large language models then clean up grammar, remove filler words, and tighten structure. Because the app is open source and provider-agnostic, you are not locked to one vendor. It is aimed at writers, office workers, and technical staff who type a lot and want a faster way to get words onto the screen.
What you can do with it
- Dictate text and have it transcribed in real time.
- Polish grammar, filler words, and structure with AI.
- Switch between many speech-to-text and LLM providers.
- Use context-aware presets for Gmail, Slack, and Docs.
- Add custom dictionaries for names and technical terms.
Who it is for
- Writers and copywriters who produce large amounts of text.
- Office workers who handle email, documents, and team chat.
- Technical staff writing code comments and documentation.
What to watch out for
The app is free and open source, but project terms and provider options can change, so check the repository for the latest. Getting a good result takes setup: pick providers, configure hotkeys, and add dictionaries for your field. Accuracy depends on your microphone and the provider you choose, so noisy audio or a weak model will show in the text. It runs as a desktop app, so you need to install it and grant the permissions it asks for. Test it with your own voice and apps before you rely on it daily.
FAQ
Is OpenTypeless free?
Yes. It is completely free and open source, though project terms can change, so check the repository for updates.
Which providers does it support?
It works with speech-to-text providers like Deepgram Nova 3, AssemblyAI, and GLM ASR, plus multiple LLM providers for polishing.
How many languages does it handle?
Up to 99 languages, with auto-detection or a target list of up to five languages.
Why it is free
What the official page says： completely free
Free limits, trial length and features are decided by the official site and can change at any time.
Product preview
Screenshot archived from the listing page, for preview only.
