Show HN: Open-source customizable AI voice dictation built on Pipecat
a day ago
- #AI-dictation
- #open-source
- #voice-interface
- Tambourine is a universal voice interface that works across various applications, allowing users to dictate text naturally.
- It offers AI-powered speech-to-text with formatting, removing filler words and adding punctuation.
- Customizable with personal dictionaries and supports multiple AI providers for STT and LLM.
- Available on multiple platforms including Windows, macOS, Android, and iOS (Linux support pending).
- Features include dual-mode recording, real-time transcription, and context-aware formatting.
- Open-source alternative to proprietary tools like Wispr Flow and Superwhisper.
- Built with Tauri for the desktop app and Pipecat for the voice AI pipeline.
- Requires microphone and accessibility permissions for full functionality.
- Supports various AI providers with generous free tiers, such as Cartesia and Cerebras.
- Development is ongoing, with core features functional but subject to changes.