Meetily
Meetily is a privacy-first AI meeting assistant that runs entirely on your own machine instead of sending your audio and transcripts to a vendor's servers. It captures a call, transcribes it in real time, and writes summaries and action items using open models you can point at Ollama or a local Whisper/Parakeet pipeline. The desktop app targets macOS and Windows, and because everything is processed locally it keeps working offline.
Key Features
- Local real-time transcription: Uses an optimized local pipeline that the project claims is about 4x faster than a default setup, built around Parakeet or Whisper models.
- Speaker diarization: Separates who said what and labels turns so summaries attribute statements correctly (part of the paid PRO plan).
- Ollama integration: Runs summarization and notes on local model endpoints, avoiding per-token API charges.
- Privacy-first by default: Audio and text stay on device; no transcript is uploaded to a cloud service.
- Offline-friendly: Works without a network connection once models are installed.
- Native desktop app: Installs as a local app on macOS and Windows rather than as a browser-based service.
Use Cases
Who Should Use This Tool?
- Compliance-sensitive teams: Legal, medical, and finance teams that cannot send meeting audio to a third party.
- Remote and hybrid workers: People who want searchable meeting notes without paying per-minute transcription fees.
- Self-hosting proponents: Users who already run Ollama or local Whisper and want their workflow to reuse it.
Problems It Solves
- Transcript privacy: Keeps sensitive discussions off cloud servers.
- Recurring transcription cost: Uses open local models instead of metered cloud speech-to-text.
- Lost follow-ups: Produces searchable summaries and action items so commitments are not forgotten.
Pricing
| Plan | Cost | Features |
|---|---|---|
| Community (CE) | $0 | Open source; local transcription, summaries, and notes with no cloud upload. |
| Meetily PRO | Paid | Enhanced accuracy, advanced exports, custom summary workflows, team-ready features, and speaker diarization. |
| Compute | You pay | Runs on your own hardware; open models used for summarization. |
A launch discount code (LAUNCH20) has offered 20% off PRO. Check the pricing page for the current offer.
Advantages & Unique Selling Points
- No cloud dependency: Everything runs locally, which is rare among polished meeting assistants.
- Open source and auditable: The MIT-licensed community edition can be self-hosted and modified.
- Uses models you already run: Ollama and local Whisper/Parakeet support fits into an existing local-AI stack.
Getting Started
- Download the release for macOS or Windows from the GitHub releases page.
- Install a local model such as Whisper or Parakeet, or point the app at an Ollama endpoint.
- Join a meeting as usual and let Meetily capture, transcribe, and summarize it locally.
- Export the notes or connect the workflow to your own docs and task boards.
Frequently Asked Questions
Does Meetily upload my audio?
No. In the community edition, transcription and summarization run locally. Review the privacy settings before enabling any optional cloud model.
Which platforms are supported?
The desktop app supports macOS and Windows.
Is speaker diarization included for free?
Speaker diarization is part of the paid PRO plan in the releases we reviewed, so verify the plan before relying on it.
Alternatives
- Whisper v3: The open speech-to-text model many local pipelines use.
- n8n: Workflow automation to route meeting notes into your own pipeline.
- Breeze TTS 2: The output side of a voice workflow, generating speech locally.
Tips
- Choose a local Whisper or Parakeet build best matching your languages to improve accuracy.
- Use Ollama for summaries so the whole meeting loop stays offline.
- Export notes to a searchable store if you want to find past decisions quickly.
Conclusion
Meetily is a fast, open-source, local-first meeting assistant that gives you private transcription and summaries without cloud uploads or per-minute fees. Download the release, point it at local models, and make meeting notes something you own.
Comments
No comments yet. Be the first to comment!
Related Tools
Muse Glimmer
developer.meta.com/ai/models/muse-glimmer
Meta's open-weight 30B agentic multimodal model distilled from Muse Spark, designed to run always-on coding and tool-use agents locally on a single consumer...
Ollama
ollama.com
Local models plus Ollama cloud. Rechecked: Free $0, Pro $20/month or $200/year, Max $100 paused, Team $25/seat (5-seat min). Not a 30-minute Llama3 tutorial.
Ling-3.0-flash
huggingface.co/inclusionAI/Ling-3.0-flash
inclusionAI MIT hybrid-linear MoE: 124B total, 5.1B active. Rechecked 2026-08-21 on Hugging Face.
Related Insights
Complete Guide to Claude Skills - 10 Essential Skills Explained
Deep dive into Claude Skills extension mechanism, detailed introduction to ten core skills and Obsidian integration to help you build an efficient AI workflow
Claudesidian: Transform Obsidian into an AI-Powered Second Brain
Discover Claudesidian, an open-source project that perfectly integrates Obsidian with Claude Code. Built-in PARA method, custom commands, and automated workflows for a complete idea-to-implementation solution.
Stop Cramming AI Assistants into Chat Boxes: Clawdbot Picked the Wrong Battlefield
Clawdbot is convenient, but putting it inside Slack or Discord was the wrong design choice from day one. Chat tools are not for operating tasks, and AI isn't for chatting.