Skip to content
RepoQuarry

FluidVoice altic-dev/FluidVoice

On-device voice dictation app for macOS that also records meetings (FluidMeet) with live captions, speaker labels and transcripts.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

FluidVoice at a glance

Checked 15 Oct 2026

FluidVoice is a macOS dictation app (macOS 15+, Apple Silicon; Intel via Whisper only) using on-device Parakeet, Nemotron, Cohere, Apple Speech or Whisper models. Since v1.6.11 (October 2026) its FluidMeet feature records meetings locally with live captions, speaker labels and transcripts; meeting summaries are only in the 1.6.12 beta. App code is GPLv3, but the optional Fluid Intelligence model is privately maintained. No Windows or Linux build yet.

Cite this pageRepoQuarry. “FluidVoice.” Checked 15 Oct 2026. https://repoquarry.dev/projects/fluidvoice

How much of Otter.ai does it replace?

3 of 3 jobs. Local meeting transcripts with speaker labels on macOS only; meeting summaries are in the 1.6.12 beta, disabled in stable 1.6.11.

Partial: Meeting transcription Partial replacement
Partial: Speaker identification Partial replacement
Partial: Meeting summaries Partial replacement

How much of Granola does it replace?

2 of 3 jobs. Records and transcribes meetings locally without a bot; summaries only in the 1.6.12 beta; no chat across meetings (a local API lets coding agents read them).

Partial: Meeting transcription Partial replacement
Partial: AI meeting notes Partial replacement
No: Chat across meetings Not covered

For chat across meetings, see other Granola alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction
QuestionAnswerSourceChecked
Runs on my machine? Confirmed: Yes, native macOS app; transcription runs on-device your voice and text never leave your machine unless you opt in to a cloud AI provider github.com · FluidVoice#features ↗ 1d ago
Code licence Partial: GPL-3.0 (copyleft) from 2026-02-23; earlier versions Apache-2.0. Fluid Intelligence runtime is private From 2026-02-23 onward, this project is licensed under the GNU General Public License, Version 3.0 (GPLv3). github.com · FluidVoice#license ↗ 1d ago
Model weights licence Partial: Mixed; Whisper MIT, Parakeet CC-BY-4.0; Nemotron, Cohere, Fluid Speech and the private Fluid Intelligence model terms not stated in the README Fluid Intelligence is a separate, privately maintained local AI runtime that powers advanced on-device dictation enhancement github.com · FluidVoice#fluid-intelligence ↗ 1d ago
Commercial use Partial: GPLv3 app allows commercial use; weights licences vary by model and are not all documented Versions published before this date were licensed under Apache License 2.0. github.com · FluidVoice#license ↗ 1d ago
Paid services needed? Partial: None required; cloud AI enhancement via your own OpenAI, Groq or custom provider key is optional (Optional) Bring your own AI provider — add an OpenAI, Groq, or custom provider API key for cloud-based enhancement. github.com · FluidVoice#quick-start ↗ 1d ago
Hardware Confirmed: Apple Silicon Mac for all models (Intel Macs via Whisper only); ~1 GB disk per voice model, ~3.5 GB for Fluid Intelligence Apple Silicon Mac for all models github.com · FluidVoice#requirements ↗ 1d ago
Operating systems Partial: macOS 15+ only; Windows and iOS announced but not released; no Linux FluidVoice is on macOS today. iOS and Windows are on the way github.com · FluidVoice#readme ↗ 1d ago
Setup effort Confirmed: Easy; Homebrew cask or DMG from Releases brew install --cask fluidvoice github.com · FluidVoice#quick-start ↗ 1d ago

Get the weekly top 5 and what’s new, every Monday

We’ll email you once a week. Unsubscribe any time. We never share your email.

Cookie preferences

Choose what we may store in your browser. You can change this any time from Cookie settings in the footer.

Essential

Security, and remembering this choice. rq-consent · 12 months

Always on