Skip to content
RepoQuarry

Build · Intermediate · A weekend

Build a meeting notes and voice-to-task assistant with open-source software

Turn meetings and spoken notes into tasks, summaries and subtitle files.

Instead of paying for Otter.ai, ChatGPT Plus, Zapier, Jira or Linear.

What you'll build

Talk, and the work gets written down: record a meeting or a voice note, and get a transcript, a short summary and a list of action items placed straight into your task tracker. The same transcription also makes subtitle files for videos, and keeps a running record of tabletop game sessions. It replaces meeting-note services like Otter.ai, with recordings that never leave your machine.

Blueprint

Meetily or whisper.cpp transcribes (whisper.cpp also writes subtitle files), Ollama pulls out the action items, and Activepieces creates them in Plane.

  1. You Meeting audio or a voice note Meetily
  2. Meetily The transcript Ollama
  3. Ollama Action items with owners and dates Activepieces
  4. Activepieces New tasks Plane

The repos

  1. 01 · Record and transcribe meetings Meetily

    Local AI meeting assistant that records mic and system audio, transcribes live with Whisper or Parakeet and summarises with an LLM.

    Setup
    Easy on Windows/macOS (installer); Linux needs a Rust and Node source build source
    Hardware
    Windows installer needs an AVX2 x64 CPU; RAM and VRAM needs not documented source
    Paid extras
    None required with Ollama; cloud LLMs and Meetily PRO are optional paid extras
    ★ 32k MIT Confirmed: 8/8 facts Official docsGitHubReplaces Otter.ai
  2. 02 · Transcribe and subtitle whisper.cpp

    Dependency-free C/C++ port of OpenAI's Whisper speech recognition, running on CPU or GPU across desktop, mobile and the browser.

    Setup
    Hard; build from source with CMake, then download a ggml model source
    Hardware
    CPU fine; memory from ~273 MB (tiny) to ~3.9 GB (large) source
    Paid extras
    None required
    ★ 55k MIT Confirmed: 8/8 facts Official docsGitHubReplaces Otter.ai
  3. 03 · Pull out actions Ollama

    Local model runner that downloads and serves open LLMs (Gemma, Qwen, DeepSeek, gpt-oss and more) with a CLI and local API.

    Setup
    Easy; one-line install script or installer, then ollama run source
    Hardware
    Apple M-series or x86 (CPU only) on Mac; 4GB for Windows binary; RAM not stated source
    Paid extras
    None for local use; Ollama Cloud plans optional (Pro $20/month)
    ★ 184k MIT Confirmed: 8/8 facts Official docsGitHubReplaces ChatGPT Plus
  4. 04 · Send to the task list Activepieces

    Self-hostable no-code automation builder with 280+ TypeScript integrations ("pieces") that are also exposed as MCP servers.

    Setup
    Medium, install script or a single docker run with embedded database source
    Hardware
    Not documented
    Paid extras
    None required; enterprise licence and Activepieces Cloud are optional
    ★ 25k MIT Partial: 7/8 facts Official docsGitHubReplaces Zapier
  5. 05 · Track the tasks Plane

    Self-hostable project management with work items (issues), cycles (sprints), modules, saved views, pages and analytics; also sold as Plane Cloud.

    Setup
    Medium; install script with Docker Compose, or Kubernetes source
    Hardware
    2 CPU cores, 4 GB RAM (8 GB recommended for production) source
    Paid extras
    None required; paid licence keys and Plane Cloud are optional
    ★ 61k AGPL-3.0 Confirmed: 8/8 facts Official docsGitHubReplaces Jira, Linear

Step by step

Install commands change, so each step links to the repo's own docs for the exact commands.

  1. Start recording meetings

    Install Meetily on the computer you take calls on. It records and transcribes locally; tell people you're recording.

    Uses Meetily · its docs

  2. Or transcribe files with whisper.cpp

    For voice notes and videos, whisper.cpp turns audio files into text and can write subtitle files. It's built from source, so it's the harder install.

    Uses whisper.cpp · its docs

  3. Pull out the actions

    Send the transcript to an Ollama model and ask for a short summary plus a list of action items with who owns each and any date mentioned.

    Uses Ollama · its docs

  4. Create the tasks

    Build an Activepieces flow that takes that list and creates a task in Plane for each item.

    Uses Activepieces · its docs

  5. Review once a day

    Check new tasks in Plane each morning and delete or fix the ones the model got wrong.

    Uses Plane · its docs

Take it further

  • Subtitle and translate your videos by transcribing, translating with Ollama, and saving the subtitle file.
  • A game-night lore keeper that tracks characters, places and loot across sessions.
  • Send spoken to-dos from your phone into the same flow.

Shared builds

Nobody has shared one yet. Be the first.

Built your own version and put it on GitHub as open source? Share your build

Questions

Do I need to tell people I'm recording?

Yes. Recording laws differ, and many places require everyone's consent. Say so at the start.

Meetily or whisper.cpp?

Meetily for live meetings on your computer; whisper.cpp for audio and video files and subtitles.

Can it use Trello or Jira instead of Plane?

Yes; Activepieces connects to many tools. Plane is here because it's open source.

Every repo here is open source, still maintained and checked against its own documentation. How we check →

Cookie preferences

Choose what we may store in your browser. You can change this any time from Cookie settings in the footer.

Essential

Security, and remembering this choice. rq-consent · 12 months

Always on