Overview
This section tracks local-first AI applications and tooling: software that runs models entirely on the user’s own hardware rather than calling a cloud API. The current focus is local voice AI (TTS/STT/voice cloning), with room to expand into local chat/LLM front-ends, image generation studios, and the inference runtimes underneath them (Ollama, llama.cpp, LM Studio) as entries are added.
Entries
- Voicebox — open-source local voice AI studio: 7-engine TTS voice cloning, Whisper-based dictation, and an MCP server that gives AI coding agents a spoken voice