Download FIA

v0.1.0 — first public release
Windows 11 / Windows 10 (2004+) · ~20 MB installer
Download installer
Windows SmartScreen: the installer is not yet code-signed, so Windows will show a warning. Click More info → Run anyway to proceed. This is expected for unsigned builds and will be resolved once a signing certificate is in place.

System requirements

OS: Windows 11, or Windows 10 version 2004+
RAM: 8 GB recommended (Whisper models)
Microphone: any — for your side of the call
Network: required on first run (model download) and for the LLM API
API key: BgGPT (free tier OK) or any OpenAI-compatible endpoint
Phone calls: optional — Windows Phone Link + Bluetooth

First-run setup

  1. Install and launch

    The speech model downloads once on first start (a few hundred MB, depending on the size you choose — small or medium).

  2. Enter your API key

    Settings → LLM → API key. BgGPT is preconfigured; any OpenAI-compatible base URL works.

  3. Choose your notes folder

    Point FIA at the directory with your documents. Each subfolder can hold a бележки за интервю.md file — use the included template.

  4. Pick your response language

    The hint language is independent of the speech language — listen in Bulgarian, get hints in English, or vice versa.

Release notes

v0.1.0

First public build: local Whisper transcription, WASAPI loopback capture, question gate, document-grounded hints (local retrieval over your notes), private always-on-top overlay, streaming OpenAI-compatible LLM support, startup version check.

All releases are hosted on GitHub. Older builds may be disabled remotely when a critical update ships — the app will tell you and link here.