LLM-Minutes-of-Meeting
Turn audio and video recordings into structured meeting minutes and action items.
Convert recordings into transcripts, decisions, and action items locally or via cloud APIs.
MoM AI processes audio and video recordings to generate structured transcripts, decisions, and action items. It suits professionals handling sensitive meetings who want to retain absolute control over where their data goes.
What it is
The application ingests meeting recordings, transcribes them using your choice of local or cloud speech-to-text engines, and passes the text to an OpenAI-compatible LLM for summarization. It handles arbitrarily long recordings by splitting files and merging outputs, tracking progress via Celery and Redis background workers.
Who it's for
It suits technical professionals and internal IT teams who want an in-house alternative to commercial meeting bots and value data privacy. It does not suit non-technical users who expect a managed cloud SaaS without touching Docker or command-line tools.
Key features
Flexible Transcription Engine
Choose between Deepgram, AssemblyAI, ElevenLabs Scribe, Sarvam AI, OpenAI-compatible endpoints, or local Whisper.
Universal LLM Support
Connect any OpenAI-compatible chat API including OpenAI, Anthropic, Gemini, Ollama, LM Studio, or vLLM.
Long Recording Chunking
Long audio files are transcribed and summarized in parts and then merged automatically to bypass context window limits.
Local Data Encryption
Database, logs, uploaded files, and encrypted API keys are stored locally in a single application directory.
Up and running in 3 steps
Our honest verdict
What's great
- Complete control over data storage with zero telemetry or tracking
- Switch freely between local models and cloud transcription or LLM providers
- Built-in Celery background job queue with live progress, cancellation, and retries
- Health and readiness endpoints alongside Prometheus monitoring metrics
Where it falls short
- Requires technical competency to deploy via Docker or configure Redis and Celery manually
- Running local Whisper or local LLMs demands capable hardware resources
- Cloud transcription and LLM providers incur their own separate API usage costs
Pricing & license
The software is free under its MIT license. The only costs are running your own hardware infrastructure or paying third-party fees to cloud transcription and LLM providers.
Open remote AI-training roles
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Can I run this completely offline without API keys?
Yes, by building with local Whisper and pointing the LLM to a local instance of Ollama or LM Studio.
How are API keys stored?
They are encrypted at rest using a key generated during the first run and stored locally within your application data directory.
What happens if a meeting recording is very long?
The application splits long files automatically, processes them in chunks, and merges the resulting transcripts and summaries.
