OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Chat & Assistants

Humla

Humla is an open-source macOS meeting-notes app that records microphone and system audio, transcribes conversations, identifies speakers, and creates summaries from the transcript and a user’s notes. It also offers question answering with cited note sources and a read-only MCP server for agent access.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Humla — OpenViST project profile in Chat & Assistants.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Humla when you want to keep a meeting record that combines your own notes with captured audio, then revisit it through transcripts, summaries or cited question-answering. It is also relevant when you want to choose local or hosted processing separately for transcription, summaries and chat.

Who it is for

  • Mac users who want transcripts and structured summaries alongside their own meeting notes.
  • People who need speaker-labelled records of remote or in-person conversations.
  • Users who want to search meeting notes or ask questions grounded in those notes.
  • Agent users who want read-only access to their meeting notes through MCP.

Documented uses

  • Capture microphone and computer audio as separate streams during a meeting.
  • Transcribe recorded conversations using on-device Whisper or a configured hosted provider.
  • Identify and rename speakers in a transcript.
  • Create a summary that draws on both typed notes and transcript content.
  • Ask questions about notes and inspect cited sources.
Before you choose

Requirements & limits

The README describes macOS 13 or later, recommends Apple Silicon, and links to a macOS release download. It describes the app as free and local-first, with optional paid cloud sync; transcription and summaries use the user’s chosen providers or on-device options.

  • Local Whisper runs on the Mac after a one-time model download; hosted transcription requires a provider API key.
  • Humla supports per-language transcription provider choices and a default provider.
  • Speaker identification runs on-device; the README lists Nemotron 3 and Community-1 engines.
  • Summaries and chat can use local LLM servers including Ollama, LM Studio, llama.cpp or vLLM, or supported hosted services.
  • The built-in MCP server provides read-only tools for searching and reading notes.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up fully on-device summaries and chat

Configure local language-model servers for summaries and note-question answering.

Before you start

  • Humla installed on macOS.
  • A local LLM server such as Ollama, LM Studio, or llama.cpp.

Steps

  1. Configure the local LLM server under Settings → Summary to use it for on-device summaries.

  2. Configure the local LLM server under Settings → Chat to use it for on-device chat.

Expected outcome

Humla is configured to use a local LLM for summaries and chat.

Guide limitations

  • This setup guide does not cover installing or configuring the local LLM server itself.
  • The README describes local summarization and chat separately, under their respective settings sections.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
michaelwilhelmsen/humla · Official homepage title · retained in its source language
How we review sourcesSuggest a correction