SpeakoFlow is an open-source desktop voice tool for real-time dictation into any app and voice-triggered generation of emails or replies. Speech recognition uses whisper.cpp or Parakeet, the local assistant can run through llama.cpp, Ollama, or LM Studio, and local speech output uses Kokoro, with optional user-selected cloud providers.
Features
- Real-time dictation into any app
- Floating voice assistant
- Local STT, LLM, and TTS
- Memory, search, and translation
Use Cases
- Offline voice input
- Writing emails by voice
- Screen-content Q&A
- Accessible desktop operation
FAQ
SpeakoFlow is an open-source desktop voice tool for real-time dictation into any app and voice-triggered generation of emails or replies. Speech recognition uses whisper.cpp or Parakeet, the local assistant can run through llama.cpp, Ollama, or LM Studio, and local speech output uses Kokoro, with optional user-selected cloud providers. Core capabilities include: Real-time dictation into any app, Floating voice assistant, Local STT, LLM, and TTS.
Common scenarios include: Offline voice input, Writing emails by voice, Screen-content Q&A.
