stenolabs/stenoai

StenographAI is the secure privacy-first AI notepad & notetaker for confidential conversations in government & defence sectors. On Windows, Linux & MacOS.

What it solves

StenographAI is a privacy-first AI notepad designed for confidential conversations. It allows users to record, transcribe, and summarize meetings locally on their device, ensuring that sensitive data—such as government, defense, or C-suite communications—never leaves the user's premises.

How it works

The application captures system audio and microphone input, using local AI models for transcription and summarization. It employs Parakeet TDT v3 (via MLX on Apple Silicon) for live transcription and speaker labeling, and Whisper for high-accuracy post-recording transcription across 99 languages. For summarization, it bundles Ollama to run models like Gemma 4 or Qwen 3.5 locally. Users can also optionally connect to cloud-based LLMs (OpenAI, Anthropic, AWS Bedrock) if they prefer.

Who it’s for

It is built for professionals handling highly confidential data, including those in government, defense, and executive leadership, as well as anyone who prefers a local-first AI workflow for meeting productivity.

Highlights

  • Local-First Privacy: Processes recordings, transcripts, and summaries on-device by default.
  • Live Speaker Attribution: Real-time on-screen text with [You] vs [Others] labeling.
  • [macOS] Individual Speaker Separation: Post-recording voice separation to identify different speakers.
  • System Audio Capture: Records both sides of virtual meetings via native Core Audio Tap (macOS), loopback (Windows), or PipeWire (Linux).
  • AI-Powered Querying: Ability to ask questions about specific meetings or all notes using local or cloud LLMs.
  • Agent Integration: Includes skills for AI agents (like Claude Code or Codex) to access and act upon local meeting notes.
  • Custom Report Templates: Define and generate structured summaries based on custom styles.
  • Multi-language Support: Live transcription for 25 European languages and post-stop transcription for 99 languages.

Related

  • Project
    fastrepl/anarlog

    anarlog is an open‑source, local‑first meeting‑note app that records audio on your device, transcribes it with on‑device or user‑chosen STT models, and lets you generate AI summaries or chat using any LLM you control. All notes and recordings stay in a local SQLite database (exportable to Markdown) unless you opt‑in to encrypted CloudSync. Built with Tauri (Rust backend) and React/TypeScript, it runs on macOS, Windows, Linux, and has mobile/watch companions. MIT‑licensed for the community code, with optional commercial enterprise components.

  • Project
    harry0703/AudioNotes

    A local-first audio and video transcription tool that converts recordings into structured Markdown notes and enables AI-powered Q&A over the content.

  • Project
    Zackriya-Solutions/meetily

    A privacy-first AI meeting assistant that captures, transcribes, and summarizes meetings locally on your own infrastructure to ensure data sovereignty.

  • Project
    FutureUniant/WorkShadow

    A local-first desktop work log application that uses AI for semantic search, automated report generation, and Q&A over personal work history while keeping data on the user's machine.

  • Project
    jankeesvw/omarchy-meeting-recorder

    Meeting Recorder is a Rust‑based Linux app that records microphone and system audio, transcribes locally with Whisper, provides speaker diarisation, optional AI‑generated chapters, and integrates tightly with the Omarchy desktop environment—all while keeping data on‑device.