Zackriya Solutions Releases Meetily, a Rust-Based Self-Hosted Local AI Meeting Assistant
Meetily offers local meeting transcription and summarization on macOS and Windows without cloud dependencies. Built with Rust, it optimizes Whisper and Ollama to achieve up to four times faster speech-to-text processing while ensuring absolute data privacy.
Zackriya Solutions has launched Meetily, a fully self-hosted, privacy-focused AI meeting assistant built in Rust. Designed for organizations demanding strict data sovereignty, Meetily handles audio recording, real-time transcription, and automated summarization locally without transmitting any data to external cloud services. The application achieves significant performance improvements by optimizing Parakeet and Whisper engines within its Rust-based architecture. These optimizations allow Meetily to execute live transcription up to four times faster than standard local configurations. Additionally, the tool integrates speaker diarization and leverages local large language models through Ollama to generate meeting summaries directly on the host machine. Currently available as an open-source desktop application for macOS and Windows, Meetily targets enterprises seeking to mitigate data leakage risks associated with third-party SaaS alternatives. However, because all computational workloads run locally, deployment requires systems equipped with sufficient CPU and GPU resources to maintain real-time performance.
Primary source
Open GitHub Trending ↗This article summarizes the source. Check the original for full details and conditions.
Comparison
| Aspect | Before / Alternative | After / This |
|---|---|---|
| Data Sovereignty | Audio and transcripts sent to third-party cloud servers | 100% local processing with zero external data transmission |
| Transcription Speed | Standard Whisper local implementations with potential latency | Up to 4x faster live transcription optimized via Rust and Parakeet |
| Summarization Engine | Proprietary cloud APIs with subscription costs | Local offline LLMs run via Ollama integration |
| Operating Costs | Recurring per-user licensing fees for SaaS platforms | Free and open-source, limited only by local hardware capabilities |