offmute
CLI tool that transcribes, diarizes, and generates reports from meeting recordings using Gemini models.
Extract speaker-tagged transcripts, visual demo context, and structured summaries via the command line.
offmute is an open-source CLI tool and Node.js package that processes meeting recordings into speaker-identified transcripts and structured reports. It relies directly on Google's Gemini API to analyze audio chunks and video screenshots. Project managers and Agile teams can use it to turn recorded standups, sprint reviews, or client calls into Markdown notes.
What it is
The tool splits meeting audio into segments and extracts key video frames to capture visual demos alongside conversation. It sends these inputs to Google Gemini to identify speaker roles and build reports using a multi-stage process. Outputs are written incrementally as Markdown files so you can inspect progress while it runs.
Who it's for
It suits technical project managers, Scrum masters, and developers who want detailed meeting notes without paying per-minute SaaS transcription fees. It is not suitable for non-technical users who need a graphical interface or teams prohibited from sending meeting media to Google's cloud services.
Key features
Visual Video Analysis
Extracts screenshots at key timestamps to detect screen shares and product demos during video meetings.
Spreadfill Report Generation
Creates an initial report outline and fills each section independently to maintain context across long calls.
Smart Diarization
Identifies distinct speakers and infers their names and roles directly from speech context.
Custom Context Prompts
Accepts custom instructions to direct the model toward technical decisions, action items, or tone.
Up and running in 3 steps
Our honest verdict
What's great
- Analyzes both visual video frames and audio chunks for contextual summaries.
- Uses incremental file writes so you can read transcription progress during processing.
- Offers clear model tier choices between Gemini 2.5 Pro, Flash, and Flash Lite.
- Can be executed instantly via npx or integrated programmatically as an npm package.
Where it falls short
- Requires a local installation of ffmpeg and your own Google Gemini API key.
- Processing long video recordings can incur noticeable Google API usage costs.
- CLI-only execution lacks a graphical interface for non-developer team members.
- Transmits meeting audio and screenshots to third-party Google cloud endpoints.
Pricing & license
offmute is free and open-source under the Apache-2.0 license. The primary running cost comes from your own Google Gemini API token usage.
Open remote AI-training roles
879 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 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
Do I need a paid API key to run offmute?
You need a Google Gemini API key. Depending on your usage tier and meeting lengths, Google may charge API fees based on token consumption.
Does offmute process everything locally?
No. While media extraction and chunking happen locally via ffmpeg, all audio chunks and video screenshots are sent to Google's Gemini API for processing.
Can offmute analyze screen shares during sprint reviews?
Yes. The tool extracts video screenshots alongside audio, allowing Gemini to recognize visual presentations and product demonstrations.
