Curated open source apps, developer tools and self-hosted projects.
FluidAudio is a Swift SDK for local, low-latency audio AI on Apple devices, leveraging the Apple Neural Engine. It supports speech-to-text, text-to-speech, voice activity detection, and speaker diarization using open-source CoreML models.
EnviousWispr is a free, open-source macOS dictation app that runs Whisper and Parakeet speech-to-text models entirely on-device on Apple Silicon, with optional on-device or bring-your-own-key AI polish, snippets, custom vocabulary, and auto-paste.
OpenWhispr is a privacy-first, cross-platform voice-to-text dictation app for macOS, Windows and Linux. It supports local speech models (Whisper, NVIDIA Parakeet) or cloud providers via BYOK, plus AI agents, meeting transcription, notes and an MCP server.
