Tambourine

Tambourine is an open-source voice-to-text interface that lets you dictate directly into any application by pressing a hotkey and speaking. It uses customizable AI models for speech-to-text and language processing, removing filler words, adding punctuation, and formatting output to match your needs. You can choose from multiple STT and LLM providers, run fully local with Whisper and Ollama, or use cloud services.

The platform is built for anyone who wants to capture ideas faster than typing allows—speaking averages 130-160 words per minute versus 40-50 for typing. Features include a personal dictionary for technical terms and proper nouns, backtrack corrections (say "actually" to fix mistakes), smart list formatting, and full control over prompts and logic. Works on Windows and macOS.