Meetily

Overview

Meetily is a privacy-first AI meeting assistant that provides live transcription, speaker diarization, and meeting summarization. With over 23,000 GitHub stars, it has established itself as a leading open-source solution for intelligent meeting capture. Unlike cloud-dependent services, Meetily prioritizes local processing with support for Ollama-based summarization, ensuring your meeting data stays on your devices.

The application is built in Rust for performance and reliability, offering real-time transcription capabilities that rival commercial solutions while maintaining complete data privacy.

Features

• Live transcription: Real-time speech-to-text during meetings
• Speaker diarization: Automatic identification and labeling of speakers
• Ollama summarization: Local AI summarization using open-source models
• Privacy-first design: All processing happens locally by default
• Cross-platform: Built in Rust for wide OS compatibility
• Offline capable: No internet connection required for core functionality
• Export options: Multiple formats for transcription and summary export
• Low resource usage: Efficient Rust implementation for smooth operation

Installation

# Download the latest release from GitHub
# Available for Windows, macOS, and Linux

# Or build from source (requires Rust)
git clone https://github.com/Zackriya-Solutions/meetily
cd meetily
cargo build --release

# Run the application
./target/release/meetily

Quick Start

# Start a new meeting transcription
meetily start --output meeting-transcript.txt

# Generate a summary using Ollama
meetily summarize --input meeting-transcript.txt --model llama3

# Export to multiple formats
meetily export --input meeting-transcript.txt --format md,json

Core Concepts

Privacy-First Architecture

All transcription and summarization can run locally, meaning your meeting content never leaves your machine. This is critical for confidential business meetings.

Speaker Diarization

The system automatically identifies who is speaking and labels segments accordingly, enabling clear meeting notes that distinguish between participants.

Ollama Integration

Integration with Ollama allows powerful local summarization using models like Llama 3, eliminating the need for cloud API calls.

Advanced Features

Custom Model Configuration

Configure which Ollama model to use for summarization, balancing quality and speed based on your hardware capabilities.

Multi-Microphone Support

Capture audio from multiple sources for hybrid meetings with participants in different locations.

Scheduled Meetings

Configure automatic transcription start times for recurring meetings.

Examples

Basic Meeting Transcription

# Start transcription with default settings
meetily start --output my-meeting.txt

# Stop and generate summary
meetily stop
meetily summarize --input my-meeting.txt

Advanced Configuration

# Use a specific Ollama model for summarization
meetily summarize --input meeting.txt --model mistral --temperature 0.3

# Export with timestamps
meetily export --input meeting.txt --format md --with-timestamps

Pros

• ✅ Complete privacy with local processing
• ✅ Live real-time transcription
• ✅ Automatic speaker identification
• ✅ Ollama integration for local summarization
• ✅ Cross-platform Rust implementation
• ✅ Offline capable
• ✅ 23,000+ stars with active development

Cons

• ✅ Requires sufficient local hardware for Ollama models
• ✅ Transcription accuracy depends on audio quality
• ✅ Larger local models may be slower than cloud alternatives
• ✅ No cloud sync by default (privacy-focused)

When to Use

Meetily is ideal for organizations with strict data privacy requirements, teams wanting to avoid cloud transcription services, developers building privacy-preserving meeting tools, and anyone who wants powerful meeting assistance without compromising confidentiality.

Resources

• GitHub Repository
• Ollama Documentation