altic-dev/FluidVoice

▲ 45 stars today★ 13,091⑂ 987

Fastest and only macOS Dictation app with on-device STT and custom trained AI enhancement model. Windows, iOS testflight! Android, Linux waitlist open! A local Wispr Flow alternative.

About altic-dev/FluidVoice

altic-dev/FluidVoice is an open-source project on GitHub, mainly written in Swift. Fastest and only macOS Dictation app with on-device STT and custom trained AI enhancement model. It currently holds 13,091 stars and 987 forks with 124 open issues, and was last pushed on 2026-11-16 (repository created 2025-09-21).

Project Overview

Git Homed tracks it on the Today's Trending board, currently at rank #65 with 45 new stars today.

GitHub Repository Details

Repository altic-dev/FluidVoice · default branch main · size 123353 KB · watchers 47 · source: GitHub REST API and repository README

README

FluidVoice

https://github.com/altic-dev/FluidVoice/blob/HEAD/GitHub stars https://github.com/altic-dev/FluidVoice/blob/HEAD/Sponsor FluidVoice https://github.com/altic-dev/FluidVoice/blob/HEAD/X @fluidvoiceapp
https://github.com/altic-dev/FluidVoice/blob/HEAD/Fluid Speech Pro, Mini and Pico https://github.com/altic-dev/FluidVoice/blob/HEAD/Fluid 1.1 Ripple and Splash

https://github.com/altic-dev/FluidVoice/blob/HEAD/altic-dev%2FFluidVoice | Trendshift

Free, private dictation that runs on your device. Speak into the apps you already use, clean up your words locally, and keep a searchable record of your meetings.

This repository is the macOS app, open source under GPLv3. FluidVoice is also available on other devices through the betas below.

Website · Speech models · What’s new · iPhone beta

Install with Homebrew: brew install --cask fluidvoice

Manual download: latest release

[!WARNING]
fluidvoice.org is not affiliated with, operated by, or endorsed by FluidVoice or Altic. Its claims to be our official website are unauthorized.
> Our official website is fluidvoice.com. Our official domains also include altic.dev and fluidvoice.app.
> For official downloads, use this repository's GitHub Releases.
> If fluidvoice.org appears in your search results claiming to be our official website, please report the misleading result to the search provider.

Get FluidVoice

| Platform | Status | Download or join | | --- | --- | --- | | macOS | Available · macOS 15+ | Latest release · brew install --cask fluidvoice | | iOS | TestFlight beta | Join the iPhone beta | | Windows | Beta | Get the Windows beta | | Linux | Early beta | Get the Linux beta | | Android | In progress | Join the waitlist |

Features and available models vary by platform. The features below describe the Mac app.

[!IMPORTANT]
This project is free and open source under GPLv3. If FluidVoice is useful to you, please star the repository — it helps visibility and keeps development going.

---

Support FluidVoice

If FluidVoice helps you, you can support continued development, model training, and work across platforms on GitHub Sponsors.

---

What's New in 1.6.13

Read the full 1.6.13 release notes.

Speech recognition and text cleanup

Two separate steps, both available locally:

1. Fluid Speech turns your audio into text. Choose Pro for accuracy, Mini for a smaller download, or Pico for the smallest footprint. 2. Fluid Intelligence optionally cleans up that text: remove fillers and false starts, format lists, and apply the writing style you choose for each app.

You can use dictation without cleanup, pair a speech model with local cleanup, or bring your own cloud AI provider.

Fluid Intelligence

Fluid 1.1 powers on-device dictation cleanup through Fluid Intelligence. No cloud account or API key is needed for local cleanup.

| Model | Best for | Approximate download | | --- | --- | --- | | Fluid 1.1 Ripple | Smaller, faster cleanup for lighter Macs | ~600 MB | | Fluid 1.1 Splash | More accurate cleanup, including harder corrections | ~750 MB |

Ripple is almost twice as fast as Splash, with slightly lower accuracy. It can miss harder corrections that Splash catches.

On a smaller Apple Silicon Mac, Fluid Speech Mini + Fluid 1.1 Ripple is a good starting point. Choose speech and cleanup models separately in the app; Dashboard also recommends models for your Mac. Ripple and Splash are research previews, so review their changes before sending. Optional screen context is off by default.

The macOS app is open source under GPLv3. Fluid Intelligence is a separate, privately maintained local AI runtime included in official downloads; it is not part of this repository's source build. The app also works with supported speech models and optional cloud AI providers without Fluid Intelligence.

We're keeping Fluid Intelligence private for now so we can sustainably offer the core dictation experience for free. This may change in the future.

---

FluidMeet

Record online or in-person meetings, with audio and transcription kept on your Mac.

More about FluidMeet.

Fluid Intelligence demos

Email Template Flowers
Change Time & Name Emoji
Hyphens & Numbers

Demo

Command Mode — Take any action on your Mac using FluidVoice

https://github.com/user-attachments/assets/ffb47afd-1621-432a-bdca-baa4b8526301

Edit Mode — Write or rewrite text in any text box in any app

https://github.com/user-attachments/assets/c57ef6d5-f0a1-4a3f-a121-637533442c24

---

Features

---

Supported Models

These are speech-to-text models, separate from the cleanup models above. Downloads below reflect the Mac app's model catalog; sizes can change with model updates.

| Model | Best for | Language support | Approximate download | Hardware | | --- | --- | --- | --- | --- | | Fluid Speech Pro | Our most accurate dictation model | 25 languages | 622 MiB | Apple Silicon | | Fluid Speech Mini | Small download, clear dictation in noisy places | 25 languages | 242 MiB | Apple Silicon | | Fluid Speech Pico | Smallest Fluid Speech model | 25 languages | 154 MiB | Apple Silicon | | Nemotron Speech 3.5 — Ultra Fast Low Latency | Streaming-capable multilingual dictation | ~40 languages | 668 MiB | Apple Silicon | | Nemotron 3.5 Multilingual | Multilingual dictation | ~40 languages | 531 MiB | Apple Silicon | | Parakeet Flash (Beta) | Live English dictation | English | 428 MiB | Apple Silicon | | Parakeet TDT v3 | Fast multilingual dictation | 25 languages | 461 MiB | Apple Silicon | | Parakeet TDT v2 | Fast English dictation | English | 443 MiB | Apple Silicon | | Qwen3 ASR 1.7B | Multilingual dictation with automatic language detection | 30 languages | 2.03 GiB | Apple Silicon | | Cohere Transcribe | Multilingual dictation | 14 languages | 1.54 GiB | Apple Silicon | | Apple Speech | Native macOS speech recognition | System-dependent | Built-in; system language assets may be needed | Apple Silicon + Intel | | Whisper Tiny / Base / Small / Medium / Large Turbo / Large | Broad language coverage and Intel support | Up to 99 languages | 44 MiB–1.55 GiB | Apple Silicon + Intel |

About the models: Fluid Speech Pro is a 0.63B-parameter model with open weights under CC BY 4.0. At its November 2026 launch, it ranked first among open models under 1B parameters on the Open ASR Leaderboard. See the Pro model card and benchmarks and Mini's measurements and limitations. The Mini article describes the earlier English release; the Mac app's 1.6.13 Mini and Pico models support the 25 languages below.

Fluid Speech and Parakeet TDT v3 Languages

Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, and Ukrainian.

Parakeet TDT v2 Languages

English.

Cohere Transcribe Languages

English, French, German, Italian, Spanish, Portuguese, Greek, Dutch, Polish, Mandarin, Japanese, Korean, Vietnamese, and Arabic.

Apple Speech Languages

System language support depends on the macOS speech recognition languages available on your machine.

Whisper Language Support

Whisper supports up to 99 languages, depending on the model size you choose.

---

Quick Start

1. Install with Homebrew:

   brew install --cask fluidvoice
   
Or download the latest release.

2. Grant permissions — FluidVoice will ask for microphone and accessibility access. Both are required for dictation and typing into other apps.

3. Set your hotkey — pick a global hotkey in settings that triggers voice capture from anywhere.

4. Go through onboarding — choose your voice model based on your language and latency needs. Start with Fluid Speech Pro for accuracy, Mini or Pico for a smaller download, or Whisper on Intel. Other models cover additional languages.

5. (Optional) Enable Fluid Intelligence — choose Ripple or Splash for on-device text cleanup. No cloud account or API key is needed.

6. (Optional) Bring your own AI provider — add an OpenAI, Groq, or custom provider API key for cloud-based enhancement. Keys are stored securely in macOS Keychain. Select "Always allow" for key access.

7. (Optional) Opt in to beta builds — Settings → Automatic Updates → Beta Releases for early access to new features.

To pause listening without ending a hands-free dictation, enable Settings → Dictation → Hold Space to Mute. Start dictation with your usual shortcut, hold Space to mute, and release it to resume. The overlay shows a muted microphone. A quick tap (up to 250 ms) types one space on release; a longer hold types nothing. Muting starts immediately in both cases. Your normal shortcut still finishes the recording. Muted audio is excluded from live transcription, the final transcript, and saved audio history. Space works normally outside dictation; modifier-plus-Space and configured Space shortcuts keep their existing behavior. The option is off by default and applies to Dictation mode.

---

Requirements

---

Building from Source

git clone https://github.com/altic-dev/FluidVoice.git
cd FluidVoice
open Fluid.xcodeproj

Build and run in Xcode. All dependencies are managed via Swift Package Manager.

Run a signed Debug build using the script:

./build.sh

The signed build is written to DerivedData/Build/Products/Debug/FluidVoice Debug.app. Keep launching that product after each rebuild so macOS can preserve its Accessibility authorization.

For CI or contributors who do not have a signing identity, use the explicit unsigned fallback:

./build.sh unsigned

Unsigned builds are tied to a specific executable version and may require Accessibility, Screen Recording, and microphone permissions to be removed and granted again after rebuilding.

---

Contributing

Contributions are welcome! Read the contribution guidelines before starting work.

Development Setup

1. Clone and open in Xcode as above. 2. Signing: FluidVoice → Signing & Capabilities → Automatically manage signing → pick your Team (Personal Team is fine). If you have certificates for multiple teams, select one without changing the project by running FLUIDVOICE_DEVELOPMENT_TEAM=YOUR_TEAM_ID ./build.sh. 3. Build and run — SPM handles dependencies. 4. (Optional) Pre-commit hook to prevent accidental team ID commits:

   cp scripts/check-team-id.sh .git/hooks/pre-commit
   chmod +x .git/hooks/pre-commit
   

Pull Request Guidelines

These requirements are mandatory. PRs that do not follow the contribution guidelines will not be reviewed or merged until they comply, and may be closed.

---

Run Integration Tests

xcodebuild test -project Fluid.xcodeproj -scheme Fluid -destination 'platform=macOS'

CI uses unsigned builds:

xcodebuild test -project Fluid.xcodeproj -scheme Fluid -destination 'platform=macOS' CODE_SIGNING_REQUIRED=NO CODE_SIGNING_ALLOWED=NO

---

Privacy & Analytics

FluidVoice is local-first. Your voice, audio, and transcribed text never leave your machine unless you explicitly opt in to a cloud AI provider.

What's Collected

FluidVoice records one anonymous activity signal per local day and uploads the week's buffered signals together after the week ends. Detailed anonymous analytics are enabled by default and can be disabled at any time from Settings → Share Detailed Anonymous Analytics; when disabled, only the weekly activity batch is sent.

Daily activity:

With detailed analytics enabled: Not Collected: ---

Community

Join our Discord: https://discord.gg/VUPHaKSvYV

Follow development on X: @fluidvoiceapp

---

License

From 2026-02-23 onward, this project is licensed under the GNU General Public License, Version 3.0 (GPLv3).

Versions published before this date were licensed under Apache License 2.0.

GitHub Stars & Activity

13,091Stars
987Forks
124Open issues
SwiftLanguage

GitHub Popularity

GitHub stars13,091
Forks987
Open issues124
Primary languageSwift
LicenseGPL-3.0
Stars gained today45
Created2025-09-21
Last pushed2026-11-16

Trending History

Daily boardrank #65 · ▲ 45 stars

Related GitHub Projects

1

utmapp / UTM

Swift★ 36,918⑂ 1,921▲ 25 stars
→
2

suversal / WrapPin

Swift★ 945⑂ 139▲ 36 stars
→
3

jackfrued / Python-100-Days

Jupyter Notebook★ 187,858⑂ 55,713▲ 23 stars
→
4

microsoft / generative-ai-for-beginners

Jupyter Notebook★ 123,650⑂ 66,506▲ 99 stars
→
5

godotengine / godot

C++★ 120,261⑂ 27,533▲ 91 stars
→
6

localsend / localsend

Dart★ 99,230⑂ 5,687▲ 154 stars
→
7

louislam / uptime-kuma

JavaScript★ 94,465⑂ 8,850▲ 353 stars
→
8

unslothai / unsloth

Python★ 79,596⑂ 7,427▲ 61 stars
→

More Trending Repositories