An open-source platform for AI voice dictation that you can customize and personalize to your exact needs. Choose your models, write your prompts, make it yours.
Tambourine isn't a locked-down app. It's infrastructure you can build on to create exactly the voice experience you need.
Pick the STT and LLM that work for you. Use cloud providers or run fully local with Whisper and Ollama.
Customize how your speech gets formatted. Add your own rules, adjust the tone, define how punctuation works.
Implement features like backtracking corrections and list formatting your way. The logic is in prompts you control.
These features ship by default, but they're implemented in customizable prompts. Modify them, extend them, or build your own.
Teach it technical terms, names, and jargon specific to your work.
Say "actually" or "scratch that" to correct mistakes as you speak.
Automatically format lists.
The prompts and logic are yours to modify. Build what you need.
Tmabourine will access surrounding context to make dictation smarter.
Automatically adapt to the app you're in. Emails get salutations, code editors get syntax-aware output.
Highlight text and say "make this more formal" or "translate to Spanish" to transform it in place.
Create custom triggers like "insert meeting link" that expand to frequently used text.
Run it yourself for free, or let us handle the infrastructure.
Run the open source code yourself. Full control, completely free, requires some setup.
View on GitHubWe run the infrastructure for you. No setup, no fuss, just download and start dictating.
Join WaitlistTambourine is fully open source. Contribute features, report bugs, or fork it and build something new. The platform is yours.