Our take
Newless than six months old and changing fast- Activity
- 76
- Community
- 61
- Issues
- 21
- Pull requests
- 76
+0 stars in the last 16 days
- Merges more pull requests than 79% of the projects we track
Mityu is a fast-moving, early-stage local meeting assistant with recent releases and contributions from a broad group, but its platform coverage and validation are not yet uniform.
Good fit if
- You need local recording and transcription, and can review AI-generated notes against their source transcript.
- You use Windows and want a published installer, or are comfortable building the app from source on macOS or Linux.
- MIT-licensed software and an actively changing project suit your adoption needs.
Look elsewhere if
- You need packaged macOS support or Linux system-audio capture today.
- You require validated accuracy for noisy environments, specialized jargon, or speaker diarization; the project says its target-environment benchmark has not been performed.
- You need a tested release process before adoption; the repository reports no tests, despite frequent recent changes and releases.
All health signals
| Last commit | 2026-09-22 (17 days ago) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 11 (latest v1.2.3, 2026-09-18) |
| Contributors | 13 (top contributor: 35% of commits) |
| Issues closed, last 90 days | 0 |
| Pull requests merged, last 90 days | 62 (typically merged in 0 days) |
| Project age | 3 months |
Checked on 2026-10-10 with the GitHub API.
Overview
Mityu is a Rust and Tauri desktop app for recording meetings and conversations, transcribing them locally, and turning transcripts into summaries and action items. Its local-first design keeps recordings and transcripts on the user's machine, without requiring a cloud service or server.
It is aimed at people and teams handling sensitive conversations who want meeting notes they can check against their source transcript. Windows has a published installer; macOS and Linux users currently need to build from source, with Linux system-audio capture still unavailable.
Key Features
- Capture microphone and system audio on Windows and macOS, or microphone audio on Linux.
- Transcribe locally with Whisper or NVIDIA Parakeet, including live transcription.
- Generate summaries and action items as drafts linked to transcript segments for human review.
- Ask questions about a finished meeting and search across meetings using local transcript data.
- Separate speakers and estimate talk time after recording, using anonymous labels.
- Export approved notes to PDF, Word, or Markdown.
- Use Ollama locally or configure supported external AI services with your own API key.
Use Cases
- A privacy-conscious professional can record a meeting and review transcript-linked notes without uploading the recording to a required cloud service.
- A team can capture online calls on Windows, then approve and export summaries and action items for follow-up.
- Someone conducting in-person interviews or field conversations can record and transcribe them locally, including on Linux with microphone capture.
- A user with existing audio recordings can import and transcribe them, or re-transcribe them with a different model or language.
What you need
Detected in the repository
- Rust 1.77 or newer (from Cargo.toml)
- Automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
On Windows, download the latest x64-setup.exe from Releases and run the installer. To build from source, see the README and build guide.
Alternatives
- meetily: Meetily offers local transcription and summaries with configurable AI providers, while Mityu emphasizes source-linked notes and on-device processing.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| mityu | Rust | MIT | 1 | New |
| meetily | Rust | MIT | 31.4k | Active |
Considerations
The project is marked New and has been changing quickly, with recent commits and releases. It has contributions from multiple people, with no single contributor holding a majority share; however, the measured data shows no issues closed in the recent period, so issue-resolution activity is not yet demonstrated. The repository also reports no tests.
The Windows x64 installer is the only published platform package described in the README. macOS builds are in development, and Linux system-audio capture is not implemented. The live copilot is beta, off by default, and has not yet passed its stated check against a real model. GPU acceleration is available for some builds, but the README does not make a GPU a general runtime requirement. The project is MIT-licensed.
Found this useful?
Share it with someone who would like mityu.