Overview
MosMos is a native macOS voice workspace built by Shanghai Mosi Intelligence, the team behind the OpenMOSS research program. Instead of treating dictation and meeting notes as two separate products, it fuses them into one Fn-key-driven flow: tap Fn to dictate refined text into any Mac app, or Fn+Shift to record a multi-speaker meeting. The result is a single surface that covers the whole arc of spoken work — from a first draft dictated between meetings to a structured recap with owners and deadlines afterward.
What makes MosMos notable is context awareness. It adapts output to the active application, keeps a custom glossary so names and specialized terms stay intact, and offers three polish tiers — Basic (verbatim), Standard (grammar-smoothed), and Advanced (restructured written draft). Meeting mode separates speakers, then produces a transcript, summary, decision log, and to-do list with assignees. For developers, PMs, and operators who live in Slack, Lark, or WeChat, the app-aware dictation removes a lot of copy-paste friction. The company is angel-funded with backing from IDG Capital and MiraclePlus, and the product launched publicly in mid-September 2026. Our evaluation sees it as one of the more coherent voice-productivity launches of the quarter, with the caveat that it is still free, brand new, and commercially unproven.
Key Features
- Fn-key universal dictation — Press Fn to speak, press again to insert context-aware text into any macOS app, with Basic/Standard/Advanced polish tiers.
- Multi-speaker meeting capture — Fn+Shift records, then separates speakers and returns a transcript, summary, decisions, and action items.
- App-aware context matching — Output tone and structure adapt to the active application (chat, docs, specs, email).
- Custom glossary — A keyword library preserves names, homophones, and proprietary terms exactly as spoken.
- Local-first voice recognition — Ordinary dictation and context adaptation run on-device; history stays on the Mac.
- Optimized for chat apps — Tuned for Lark, WeChat, and similar communication tools for smoother voice typing.
Pricing
| Plan | Price | For |
|---|---|---|
| Free Mac download | $0 | macOS 15+ users; voice writing + meeting notes, no seat cost |
| Paid tiers | To be announced | Reserved in ToS; subscription or usage limits expected later |
MosMos is currently free to download for Mac. The terms reserve the right to introduce pricing and limits, so treat the free tier as an early-access offer rather than a permanent plan. Prices and future tiers are subject to change — verify on the official site.
Comparison
vs. Otter: Otter is the broader meeting-transcription service with strong web, mobile, and live-captioning reach across platforms, whereas MosMos is Mac-native and pairs dictation with meeting notes in one app. Choose Otter if you need cross-platform coverage; choose MosMos if you live on a Mac and want voice writing too.
vs. Fireflies: Fireflies leans into CRM and team-workspace integrations for recorded meetings, while MosMos focuses on the individual’s spoken workflow and local-first privacy. Pick Fireflies for sales/team pipelines; pick MosMos for personal dictation plus notes.
vs. Fathom: Fathom is a polished Zoom/Meet recorder with instant highlights, whereas MosMos adds general Fn-key dictation across any Mac app. If your need is purely recorded-call summaries, Fathom fits; if you also want to dictate drafts everywhere, MosMos is the wider tool.