What's Changed

• feat/multi language summary by @p-s-vishnu in #447
• docs: archive legacy backend setup guidance [skip ci] by @safvanatzack in #463
• fix: make analytics opt-in by default [skip ci] by @safvanatzack in #452
• fix: use TDT duration head in Parakeet greedy decoder by @lorenzojb in #459
• Fix/review 447 [skip ci] by @safvanatzack in #455
• Fix/flash attention removal for vulkan by @safvanatzack in #469
• fix(readme): update trendshift badge link by @safvanatzack in #473
• fix(frontend): stabilize BlockNote ProseMirror resolution [skip ci] by @safvanatzack in #472
• feat(tray): hide main window to tray on close instead of exiting by @passabilities in #471
• fix(tauri): prevent multiple desktop app instances [skip ci] by @safvanatzack in #476
• feat(summary): add Qwen 3.5 built-in models [skip ci] by @safvanatzack in #462
• chore: readme star history update [skip ci] by @safvanatzack in #484
• hotfix(release): Fix summary base language and macOS Dock reopen behavior [skip ci] by @safvanatzack in #498
• fix(ui): improve built-in model card responsiveness [skip ci] by @safvanatzack in #501
• Release v0.4.0 by @sujithatzackriya in #502

Installation

New Contributors

• @lorenzojb made their first contribution in #459
• @passabilities made their first contribution in #471

Full Changelog: v0.3.0...v0.4.0

Contributors

• @passabilities
• @p-s-vishnu
• @sujithatzackriya
• @safvanatzack
• @lorenzojb

What's Changed

• feat: add import audio feature for external audio files by @jeremi in #304
• fix: improve transcription quality and add parallel resampling for la... by @p-s-vishnu in #342
• Feat/import audio and retranscription by @safvanatzack in #358
• feat(beta): add beta feature gating system for import/retranscribe by @safvanatzack in #366
• Fix/pr 372 by @safvanatzack in #374
• fix: resolve remaining PR #372 review bugs by @sujithatzackriya in #375
• fix: unload transcription engine after batch jobs to free memory by @safvanatzack in #378
• fix of main before release by @safvanatzack in #381

Installation

What's Next

We're actively working on the following features for the next release:

Multi-Language Summary Generation

Support for generating meeting summaries in languages other than English. Ideal for multilingual teams — transcribe in one language and get a summary in another.

Inline Transcript Editing

Edit your transcript directly without leaving the app. Fix misheard words or formatting issues right where they appear, keeping your notes accurate and clean.

Post-Processing Mode (Retranscribe)

Option to disable live transcription during a meeting and process the full recording afterward instead. Useful for noisy environments or when accuracy matters more than real-time output.

Dark Mode

A proper dark theme across the entire interface for those late-night meetings and low-light environments.

New Contributors

• @jeremi made their first contribution in #304
• @p-s-vishnu made their first contribution in #342

Full Changelog: v0.2.1...v0.3.0

Contributors

• @jeremi
• @p-s-vishnu
• @sujithatzackriya
• @safvanatzack

What's Changed

• Fix for some Whisper models, incorrect url/size by @matbe in #320
• Fix Windows system audio device selection reset bug by @matbe in #321
• Start toast content updated by @safvanatzack in #302
• Fix Whisper model metadata and model manager behavior by @safvanatzack in #326
• Fix OpenAI compatible API connection test by @5m4u66y in #316
• Fix api key and ollama host UI persistance and sync by @safvanatzack in #329
• feat: Auto-fetch models from LLM providers by @athulchandroth in #315
• fix(ffmpeg): implement build-time bundling to eliminate runtime downloads by @safvanatzack in #325
• fix: preserve model settings and Ollama endpoint across provider switches by @safvanatzack in #345
• fix: persist model config and API key to database from meeting details by @safvanatzack in #348

Installation

New Contributors

• @matbe made their first contribution in #320
• @5m4u66y made their first contribution in #316

Full Changelog: v0.2.0...v0.2.1

Contributors

• @matbe
• @athulchandroth
• @5m4u66y
• @safvanatzack

A major update bringing custom OpenAI integration, streamlined onboarding, automatic updates, and enhanced reliability. This release represents 314 commits of improvements focused on giving you more control and flexibility while maintaining privacy-first principles.

Installation

🪟 Windows

1. Download the latest x64-setup.exe from meetily_0.2.0_x64-setup.exe
2. Right-click the downloaded file → Properties → Check Unblock → Click OK
3. Run the installer (if Windows shows a security warning: Click More info → Run anyway)

🍎 macOS

1. Download meetily_0.2.0_aarch64.dmg from meetily_0.2.0_aarch64.dmg
2. Open the downloaded .dmg file
3. Drag Meetily to your Applications folder
4. Open Meetily from Applications folder

🐧 Linux

Build from source following our detailed guides:

• Building on Linux
• General Build Instructions

🎉 Major New Features

1. Custom OpenAI Integration ⭐ (Most Awaited!)

Connect your own OpenAI-compatible endpoints for maximum flexibility and control. Use any OpenAI-compatible API service with Meetily, including custom deployments, Azure OpenAI, and third-party providers. This gives you complete control over costs, models, and API endpoints.

Addresses: #289, #274, #210, #263, #152, #232 | Discord 1, Discord 2

2. Simplified Onboarding Experience

Completely rebuilt onboarding flow with:

• 4-step guided setup - Quick and intuitive
• Background model downloads - Continue setup while models download
• Smart model recommendations - Context-aware suggestions based on your system
• Better permission handling - Clear macOS permission dialogs

Addresses: #254 | Discord 1, Discord 2, Discord 3

3. OTA (Over-The-Air) Updates

Automatic application updates mean you'll always have the latest features and fixes without manual downloads.

4. Transcript Recovery

Never lose your notes! Automatic transcript recovery protects your data if:

• App crashes during recording
• Recording is interrupted
• System restarts unexpectedly

Addresses: Discord

5. Truly Offline Mode (Built-in AI)

For situations without internet access, Meetily now includes built-in AI capabilities. While we recommend Ollama or cloud providers for best quality, the offline mode ensures you can still generate summaries when needed.

✨ Enhancements

Model Management

• Enhanced Download Progress - Real-time MB/s speed tracking with visual progress bars
• Cancel Downloads - Stop model downloads mid-way without corrupting files
• Updated Claude Models List - Latest Claude model options (Discord)
• Improved Parakeet Integration - Better local transcription model support

Audio & Recording

• FFmpeg Audio Bug Fix - Fixed critical audio corruption in release builds
• FFmpeg Path Detection - Dynamic path finding instead of hardcoded paths
• Better Error Handling - Improved audio pipeline error recovery

UI/UX Improvements

• Toast Notifications - Better feedback for user actions with stylish notifications
• Onboarding Styling - Polished, modern onboarding interface
• Visual Recording Indicator - Clear on-screen indicator when recording is active
• Auto-scroll Improvements - Fixed transcript auto-scroll behavior
• Model List During Downloads - See available models even while downloading
• Settings Modal Enhancements - Cleaner settings interface with better organization
• Summary Generation Controls - Stop generation button and better UI feedback

Permission & Platform Support

• macOS Permission Dialogs - Native permission request dialogs
• Windows Build Improvements - Fixed sidecar spawning and CMD popup issues
• Cross-platform Compatibility - Better support across all platforms

🐛 Bug Fixes

Critical Fixes

• UTF-8 String Slicing Panic - Fixed crash with multi-byte characters
• Model Selection Bug - Fixed model selection not respecting saved configuration
• Recording Metadata & Transcript Saving - Resolved data persistence issues
• Download Duplication - Fixed duplicate model download triggers
• Orphaned Processes - Proper cleanup of background processes on app exit
• Audio Device Selection Persistence - Fixed settings reverting when navigating between tabs (#237)
• Recording Settings Persistence - Auto-save recording preference now persists correctly (#234)

UI/UX Fixes

• Sidebar Start and Modal Issues - Fixed sidebar interaction bugs
• Auto-scroll Mismatch - Corrected transcript scrolling behavior
• Default Model Issue - Fixed default model selection logic
• Summary Loss on Cancellation - Preserve partial summaries when generation is stopped
• Model List While Downloading - Fixed display issues during active downloads
• Permission Check Bug - Resolved permission detection issues

Build & Platform Fixes

• Windows PowerShell Popup - Eliminated unwanted console windows
• Build Error Fixes - Resolved multiple build configuration issues
• Package Mismatch Error - Fixed dependency version conflicts
• Config Data Sync - Resolved configuration synchronization issues

🔧 Technical Improvements

Code Quality

• Component Refactoring - Extracted hooks, contexts, and services for better maintainability
• Code Cleanup - Removed unused code, variables, and functions
• Better Error Handling - Comprehensive error handling throughout the application

Developer Experience

• Updated Build Scripts - Improved build scripts for all platforms
• Dev Script Updates - Better development workflow tools
• Migration Scripts - Database migration handling
• Data Injection Script - Development testing utilities

Performance

• Model Download Performance - Optimized download speeds and progress tracking
• Token Estimation Improvements - More accurate token counting
• Onboarding Completion Order - Optimized setup flow for faster completion

📋 Other Changes

• Added timestamp to summary generation prompts
• Enhanced download error handling in UI
• Improved test endpoint functionality
• Cleanup of partial download errors
• Increased timeout for long script handling

🙏 Notes

Total Commits Since v0.1.1: 314commits

This release represents a major step forward in making Meetily more powerful, reliable, and user-friendly. The addition of offline AI capabilities and custom OpenAI integration gives you unprecedented flexibility in how you use Meetily.

Known Issues:

• Check our GitHub issues page for the latest known issues and workarounds

Upgrade Notes:

• The new onboarding will guide you through any necessary migrations
• Your existing meetings and settings will be preserved

Thank you for using Meetily!

Full Changelog: 0.1.1...v0.2.0

Release Date: October 23, 2025

Our biggest update yet! Meetily is now a single standalone application (no separate server needed!), with 3x faster transcription, pause/resume recording, system tray, template-based AI summaries, and major audio quality improvements.

📦 Installation

🪟 Windows

Direct Download:

• meetily_0.1.1.1_x64-setup.exe

Installation Steps:

1. Download meetily_0.1.1_x64-setup.exe
2. Run installer

Troubleshooting guide
If the app signature is not recognized by windows, it'll show error while installing. To solve this,
1. Right-click the .exe file → **Properties** → Check **Unblock** → **OK**

🍎 macOS (M Series chips)

1. meetily_0.1.1_aarch64.dmg]
2. Install the .dmg
3. Then open meetily.app from Applications folder.

⚠️ Data Migration: The application will ask whether to import your old database through a popup window on first launch.

🐧 Linux

Build from source: See docs/BUILDING.md and docs/building_in_linux.md

✨ What's New

🏗️ Unified Architecture (Biggest Change!)

• No separate server needed - Everything runs in one app
• Embedded SQLite database - All data stored locally
• Faster & simpler - No network latency, single executable
• Zero configuration - Just install and run

⚡ 3x Faster Transcription

• New Parakeet model for real-time speech-to-text
• Enhanced Whisper processing (35% faster)
• Automatic GPU acceleration (Metal/CUDA/Vulkan)
• Voice Activity Detection - 50% less waiting time (2s → 1.5s)

🎙️ Recording Improvements

• Pause/Resume - Take breaks without stopping
• System Tray - Control recording from menu bar
• Professional Audio - Advanced noise removal, RMS-based mixing
• Bluetooth support - Automatic resampling for all devices

🤖 Smart AI Summaries

• 7 meeting templates (Standup, Project Sync, Sales Call, etc.)
• Rich text editor (BlockNote) with formatting
• Custom Ollama endpoints - Connect your own LLM server
• OpenRouter support - Access multiple AI models
• Fully offline option - Use local Ollama, or cloud providers

🎨 Better UX

• Auto-scrolling transcripts with streaming effects
• Model management interface (download/switch models easily)
• Customizable notifications (start/stop/pause/complete)
• Permission management UI
• Improved layouts and animations

📊 Privacy-First Analytics (Opt-in)

• Help improve Meetily with usage insights
• Completely optional and transparent
• PostHog integration with full user control

🔧 Platform Improvements

Windows

• Fixed FFMPEG popup windows during operations
• Optimized WASAPI audio capture
• Improved stability and VAD tuning

macOS

• Fixed system audio echo
• Enhanced Core Audio integration
• Better Metal GPU acceleration

Linux

• Improved Vulkan compatibility
• Enhanced ALSA audio support

All Platforms

• 40% memory reduction for long recordings
• Better audio device detection
• Reduced latency and improved sync

🔄 Upgrading from v0.0.5

Good news: Your data migrates automatically!

• First launch - Migration window appears to import your existing database
• All preserved - Recordings, transcripts, and meeting data transfer seamlessly
• Homebrew users - Old data automatically detected and imported

What resets:

• Notification preferences (reconfigure in settings)
• Some models may need re-download
• Analytics preferences (opt-in required)

⚠️ Known Issues

• macOS Sequoia (15.6+): Audio capture issues? Temporarily adjust Hz to lower value, then back to 48000Hz. Details
• Linux: Some devices may need ALSA mixer configuration
• Bluetooth: Minor playback issues on older devices. See BLUETOOTH_PLAYBACK_NOTICE.md
• Templates: Minor UI polish needed for edge cases

For more information and technical details got to https://discussions.apple.com/thread/256140785
Bluetooth Devices: Some playback issues may occur with certain Bluetooth audio devices](https://discussions.apple.com/content/attachment/e07cf310-419b-45d6-a839-1e0333463281)

🙏 Thank You

Huge thanks to our community for feedback, bug reports, and contributions!

README · Report Issues · Full Changelog

This major release focuses on making Meetily easier to install and use, with significant improvements to the installation process, Docker support, enhanced UI, and privacy-focused analytics.

📦 Installation Guide

🎯 Option 1: Easiest Installation (Recommended)

Windows Users (5-10 minutes)

Step 1: Install the Frontend Application

1. Download the installer:

   • Go to Latest Releases
   • Look for and download the file ending with x64-setup.exe (e.g., meetily-frontend_0.0.5_x64-setup.exe)

2. Prepare the installer:

   • Navigate to your Downloads folder
   • Right-click the downloaded .exe file → Properties
   • At the bottom, check the Unblock checkbox → Click OK
   • This prevents Windows security warnings during installation

3. Run the installation:

   • Double-click the installer to launch it
   • If Windows shows a security warning:
     • Click More info → Run anyway, OR
     • Follow any permission dialog prompts
   • Follow the installation wizard steps
   • The app will be installed and available on your desktop/Start menu

Step 2: Install and Start the Backend
4. Download the backend:

• From the same releases page
• Download the backend zip file (e.g., meetily_backend.zip)
• Extract the zip to a folder like C:\meetily_backend\

5. Prepare backend files:

   • Open PowerShell (search for it in Start menu)
   • Navigate to your extracted backend folder:

     cd C:\meetily_backend

   • Unblock all files (Windows security requirement):

     Get-ChildItem -Path . -Recurse | Unblock-File

6. Start the backend services:

   .\start_with_output.ps1

   • This script will:
     • Guide you through Whisper model selection (recommended: base or medium)
     • Ask for language preference (default: English)
     • Download the selected model automatically
     • Start both Whisper server (port 8178) and Meeting app (port 5167)

What happens during startup:

• Model Selection: Choose from tiny (fastest, basic accuracy) to large (slowest, best accuracy)
• Language Setup: Select your preferred language for transcription
• Auto-download: Selected models are downloaded automatically (~150MB to 1.5GB depending on model)
• Service Launch: Both transcription and meeting services start automatically

✅ Success Verification:

1. Check services are running:
   • Open browser and visit http://localhost:8178 (should show Whisper API interface)
   • Visit http://localhost:5167/docs (should show Meeting app API documentation)
2. Test the application:
   • Launch Meetily from desktop/Start menu
   • Grant microphone permissions when prompted
   • You should see the main interface ready to record meetings

Alternative Installation Methods:

Option A: MSI Installer (if exe is blocked)

• Download the .msi file instead of .exe from releases
• MSI installers are typically less likely to be blocked by Windows Defender

Option B: Windows Defender Issues
If Windows Defender blocks the installer:

1. Open Windows Security → Virus & threat protection
2. Click Manage settings under Virus & threat protection settings
3. Scroll to Exclusions → Add or remove exclusions
4. Add the downloaded installer file as an exclusion
5. Run the installer manually

macOS Users (5-7 minutes)

!!! If you already have a working meetily app, please upgrade to the latest version

Installing fresh

# One command installs everything (frontend + backend)
brew tap zackriya-solutions/meetily
brew install --cask meetily
# Start the backend server
meetily-server --language en --model medium

To update existing installation:

# Update Homebrew and get latest package information
brew update
# Update to latest version
brew upgrade --cask meetily
brew upgrade meetily-backend

If you find the following error

Error: meetily-backend: SHA-256 mismatch
Expected: f6453947bfb90ecce91990f898e46f3f94d96b3eecba6176b9adea4efb989c70
 Actual: 1c79a7d7d24f166fa29bd4c9dba36ace7a3cc28a0d08699c4a18c0b8023e2252
 File: /Users/seth/Library/Caches/Homebrew/downloads/39fb3e9ff98c94bf97003752ef3bc770d455cca44b6e5a96e29a873c369429f5--meeting-minutes-main.zip
To retry an incomplete download, remove the file above.

Please refer to the instruction in
#148 (comment)

⚠️ Data Backup Warning: You are upgrading from Meetily 0.0.4 to 0.0.5. This upgrade will automatically migrate your data to a new persistent location, but it's recommended to backup your data first.

Current Data Location (Version 0.0.4):

• Database: /opt/homebrew/Cellar/meetily-backend/0.0.4/backend/meeting_minutes.db

New Persistent Location (Version 0.0.5+):

• Database: /opt/homebrew/var/meetily/meeting_minutes.db

What Happens During Upgrade:

• ✅ Your data will be automatically migrated to the new persistent location
• ✅ Data will survive future upgrades
• ✅ The old data in the Cellar directory will be cleaned up

Backup Recommendation:
The upgrade ensures data loss doesn't happen, but it's always better to backup your data before proceeding.

Common issue*

SHA Missmatch solution

Please try the following

Run

brew cleanup && brew update

Before running

brew install --cask meetily

If you still find issues, delete the cache by following

rm Users/<your_user_name>/Library/Caches/Homebrew/downloads/39fb3e9ff98c94bf97003752ef3bc770d455cca44b6e5a96e29a873c369429f5--meeting-minutes-main.zip

Then run

brew install --cask meetily

Please let me know if this worked for you.

Then open Meetily from Applications folder.

✅ Success Check: App opens and you can start recording immediately.

🐳 Option 2: Docker Installation (Easier Dependencies)

Best for: Developers, testing, or when you want automatic dependency management
Performance Note: 20-30% slower than native installation

Quick Start (5-10 minutes)

# Clone repository
git clone https://github.com/Zackriya-Solutions/meeting-minutes
cd meeting-minutes/backend
# Windows (PowerShell)
.\build-docker.ps1 cpu
.\run-docker.ps1 start -Interactive
# macOS/Linux (Bash)
./build-docker.sh cpu
./run-docker.sh start --interactive

What you get:

• Automatic dependency management
• No need to install Python, CMake, or build tools
• Consistent environment across all platforms
• Interactive model selection
• Both services running: http://localhost:8178 and http://localhost:5167

🔧 System Requirements

• Minimum: 8GB RAM, 4GB disk space
• Recommended: 16GB+ RAM, 10GB+ disk space
• Docker: 16GB+ RAM allocated to Docker

✨ Major New Features

🎨 Modern User Interface

• shadcn/ui components: Beautiful, accessible interface throughout
• Enhanced settings: Intuitive model configuration and transcript settings
• Better navigation: Improved sidebar and user interactions
• Loading states: Clear progress feedback during operations

📊 Privacy-First Analytics (New!)

• User-controlled tracking: Complete opt-in control over data collection
• GDPR compliant: Transparent privacy policy and data handling
• Analytics disabled by default: Privacy first approach
• Usage insights: Optional performance metrics and feature tracking

🐳 Complete Docker Support (New!)

• Multi-platform containers: CPU, GPU, and macOS optimized builds
• One-click deployment: Docker Compose for instant setup
• Automated scripts: Streamlined build and deployment
• Interactive setup: Guided model and language selection

🛠 Enhanced Backend

• Robust database handling: Better error recovery and data integrity
• Improved transcript processing: More reliable and efficient processing
• Schema validation: Comprehensive input validation
• Better sorting: Fixed summary generation ordering issues

📚 Installation Troubleshooting

Windows Issues

• Windows Defender blocking: Right-click installer → Properties → Unblock
• PowerShell errors: Run Set-ExecutionPolicy Bypass -Scope Process -Force
• Build failures: Use pre-built backend instead of building from source

macOS Issues

• Permission errors: Run chmod +x on script files
• App won't open: Run xattr -cr /Applications/meetily-frontend.app
• Port conflicts: Check if ports 8178/5167 are available with lsof -i :8178

Docker Issues

• Containers crashing: Increase Docker RAM to 12GB+
• Port conflicts: Run ./run-docker.sh stop to clean up
• Model download failures: Manually download with ./run-docker.sh models download base.en

🚀 What's Improved

Installation Experience

• Pre-built releases: No more complex build processes for most users
• Automated scripts: Windows PowerShell scripts handle dependencies
• Better documentation: Step-by-step guides with success checks
• Docker option: Alternative installation method with easier dependency management

Performance & Reliability

• Enhanced error handling: Better recovery from issues
• Improved processing: More reliable transcript and summary generation
• Database robustness: Better d...

Meetily v0.0.4 Pre release with Meeting History and more.

Installation Instructions (For Windows OS and Mac OS)

Windows Installation

Backend Setup

Click on the image to see installation video

Windows Backend Setup Tutorial

Step 1 : Install the dependancies (Optional)

Open powershell as administrator

Image

enter the following command to install dependencies

cd ~/Downloads
Invoke-WebRequest -Uri "https://github.com/Zackriya-Solutions/meeting-minutes/releases/download/v0.0.4/install_dependancies_for_windows.ps1" -OutFile "install_dependancies_for_windows.ps1"
Set-ExecutionPolicy Bypass -Scope Process -Force
.\install_dependancies_for_windows.ps1

==This can take up to 30 mins depending on your memory==

==Once the installation is complete, close the powershell terminal and open a new terminal==.

image

Enter the following command to download and install the backed:

cd ~/Downloads
git clone https://github.com/Zackriya-Solutions/meeting-minutes
cd meeting-minutes/backend
.\build_whisper.cmd

Might have to run the build command again if it fails

.\build_whisper.cmd

Finally, when the installation is successful, run the backend using the following command from same terminal

.\start_with_output.ps1

Troubleshooting

1. Warning - existing chocolatey installation is detected

WhatsApp Image 2025年05月27日 at 12 31 17 PM

To address this - Either use the current chocolatey version installed or remove the current one with

rm C:\ProgramData\chocolatey

2. Error - ./start_with_output.ps1 shows security error

Run after making sure the file is unblocked

Set-ExecutionPolicy Bypass -Scope Process -Force .\start_with_output.ps1

Frontend Setup

Option 1: Using the powershell (Recommended)

Open Windows powershell and enter the command

cd ~/Downloads; Invoke-WebRequest -Uri "https://github.com/Zackriya-Solutions/meeting-minutes/releases/download/v0.0.4/meetily-frontend_0.0.4_x64-setup.exe" -OutFile "meetily-setup.exe"; Unblock-File -Path "meetily-setup.exe"; Start-Process -FilePath "meetily-setup.exe" -ArgumentList "/S" -Wait; Remove-Item "meetily-setup.exe"

Once done, open meetily from desktop icon or by searching it on the windows search tool.

Option 2: Using the Setup Executable (.exe)

1. Download the meetily-frontend_0.0.4_x64-setup.exe (3.6 MB) file
2. Give permission

• Click More info and choose Run anyway, or

• Right-click on the installer (.exe) from the downloads folder, select Properties, and check the Unblock checkbox at the bottom

Windows Security Warning

3. Double-click the installer to run it

4. Follow the on-screen instructions to complete the installation

5. The application will be available on your desktop

Option 3: Using the MSI Installer (.msi)

1. Download the meetily-frontend_0.0.4_x64_en-US.msi (5.24 MB file

2. Double-click the MSI file to run it

3. Follow the installation wizard to complete the setup

4. The application will be installed and available on your desktop

macOS Installation

Frontend Setup

Option 1: Using Homebrew (Recommended)

==(Optional)==
# If meetily is already installed in your system, uninstall the current versions
brew uninstall meetily
brew uninstall meetily-backend
brew untap zackriya-solutions/meetily

 
# Install Meetily using Homebrew
brew tap zackriya-solutions/meetily
brew install --cask meetily
 
# Start the backend server
meetily-server --language en --model medium

Option 2: Manual Installation

1. Download the dmg_darwin_arch64.zip file

2. Extract the file

3. Double-click the .dmg file inside the extracted folder

4. Drag the application to your Applications folder

5. Execute the following command in terminal to remove the quarantine attribute:

xattr -c /Applications/meetily-frontend.app

Once the installation is complete, we need to grant permission once more to the updated version. To do this, remove the existing version from the system and revoke screen recording access to the old app while adding the new one.

1. Select the old app

2. Remove the app

Proceed to restart the app, start recording, and enable screen recording access.

Backend Setup

Option 1: Using Homebrew (Recommended)

# Install the backend using Homebrew
brew install meetily-backend
 
# Download a Whisper model (choose size based on your needs)
meetily-download-model medium
 
# Start the backend server
meetily-server

New Features and Improvements

We added Meeting history in this release. All your meeting data will be stored locally from now on. See the extra changes we have made :

1. Enhanced Model Configuration Management

• Added support for multiple AI model providers

• Improved model selection and configuration options

• Added OpenAI provider support for enhanced transcription capabilities

2. Windows Installation Improvements

• Fixed Windows installer bugs

• Updated Windows build instructions for Windows 10/11

• Added proper MSI installer support

• Improved installation process documentation

3. Backend Optimizations

• Streamlined backend architecture

• Removed unnecessary dependencies (ChromaDB)

• Improved process management and cleanup

• Enhanced transcript handling

4. UI/UX Updates

• Updated application logo

• Improved user interface elements

• Enhanced documentation and setup instructions

Installation Files

• Windows:

• meetily-frontend_0.0.4_x64-setup.exe (3.6 MB)

• meetily-frontend_0.0.4_x64_en-US.msi (5.24 MB)

• macOS:

• dmg_darwin_arch64.zip (8.02 MB)

System Requirements

• Windows: Windows 10/11 (64-bit)

• macOS: macOS 10.15 or later

• Hardware: Minimum 4GB RAM, 20GB free disk space

Known Issues

• The processing after recording takes too much time on windows os

Support

For support, please visit:

• Discord Channel

• Website

• GitHub Issues

Related Issues and Future Improvements

Features Implemented in v0.0.4

1. Enhanced Model Configuration Management (#77)

• Added OpenAI provider support

• Improved model selection options

• Backend cleanup and CLI testing script

2. Windows Installation Improvements (#54, #70)

• Fixed Windows installer bugs

• Addressed security warnings

• Improved installation process

3. Backend Optimizations (#64)

• Streamlined backend architecture

• Removed unnecessary dependencies

• Enhanced transcript handling

4. Meeting History and Management (#43)

• Implemented meeting history tracking

• Added management interface

• Enhanced user experience

Planned Improvements

1. Audio and Recording (#81, #68)

• Fix audio transmission issues

• Consider CosyVoice integration

• Improve app-specific audio capture

2. Platform Support (#78, #66)

• Add macOS Intel support

• Fix ARM build issues

• Improve cross-platform compatibility

3. Installation and Deployment (#74, #48)

• Fix Teams dependency issues

• Improve Homebrew Cask support

• Enhance installation process

4. UI/UX Enhancements (#62)

• Fix settings screen errors

• Improve user interface

• Add meeting history management

5. Backend Improvements (#59, #56)

• Local Ollama integration

• API interface improvements

Future Roadmap

• Linux support (#33)

• Local LLM endpoints integration (#34)

• Small model compatibility improvements (#37)

• Voice-based attendee identification (#57)

• Windows ARM support (#55)

For more details and to track progress, please visit our GitHub Issues.

What's Changed

• Fix Microphone Permission Dialog on macOS by @sujithatzackriya in #38

• Fix/audioplayer by @sujithatzackriya in #39

• Enhance/windows support by @sujithatzackriya in #44

• Introduce O...

Contributors

• @sujithatzackriya
• @athulchandroth
• @ebinzack15

Pre Release 0.0.3: Added Windows Support

A new release is available!

What's New

• Windows Support: Fixed audio capture issues on Windows
• Improved Error Handling: Better error handling and logging for audio devices
• Enhanced Device Detection: More robust audio device detection across platforms
• Windows Installers: Added both .exe and .msi installers for Windows
• Transcription quality is improved
• Bug fixes and improvements for frontend
• Better backend app build process
• Improved documentation

What would be next?

• Database connection to save meeting minutes
• Improve summarization quality for smaller LLM models
• Add download options for meeting transcriptions
• Add download option for summary

Known issues

• Smaller LLMs can hallucinate, making summarization quality poor; Please use model above 32B parameter size
• Backend build process requires CMake, C++ compiler, etc. Making it harder to build
• Backend build process requires Python 3.10 or newer
• Frontend build process requires Node.js

How to Get Started

Frontend

Windows Installation

Once the installers are downloaded from the above links [## Files for test](## Files for test), double-click the executable file to run it.

Windows will ask if you want to run untrusted apps, click More info and choose run anyway

Windows protected your PC
Microsoft Defender SmartScreen prevented an unrecognized app from starting. Running this app might put your PC at risk.
[More info](about:blank#)

To bypass this warning in windows OS,

After downloading the installer

• Right click on installer(.exe)
• Select properties option.
• Click on checkbox to check Unblock at the bottom of Properties.

windows_signing_issue

Windows frontend Installation

Option 1: Using the Setup Executable (.exe) (Recommended)

1. Download the meeting-minutes-frontend_0.1.0_x64-setup.exe file
2. Double-click the installer to run it
3. Follow the on-screen instructions to complete the installation
4. The application will be available on your desktop

Option 2: Using the MSI Installer (.msi)

1. Download the meeting-minutes-frontend_0.1.0_x64_en-US.msi file
2. Double-click the MSI file to run it
3. Follow the installation wizard to complete the setup
4. The application will be installed and available on your desktop

macOS full Installation

Option 1: Using Homebrew (Recommended)

macOS users can now enjoy a streamlined installation process with Homebrew:

# Install Meetily using Homebrew (macOS only)
brew tap zackriya-solutions/meetily
brew install --cask meetily
# Start the backend server with your preferred language
meetily-server --language en --model medium

🌐 Multi-Language Transcription Support

Our backend now supports transcription in over 90 languages! Specify your preferred language when starting the server:

# English (default)
meetily-server --language en --model medium
# Spanish
meetily-server --language es --model medium
# French
meetily-server --language fr --model medium
# German
meetily-server --language de --model medium
# Japanese
meetily-server --language ja --model medium
# Chinese
meetily-server --language zh --model medium
# Auto-detect language (may be less accurate)
meetily-server --language all --model medium

🔄 Model Selection for macOS

Choose from various Whisper models based on your needs:

# Download your preferred model
meetily-download-model medium
# Available options:
# - tiny (fastest, least accurate)
# - base (fast, moderate accuracy)
# - small (balanced performance)
# - medium (recommended for most users)
# - large-v3 (highest accuracy, requires more resources)

✨ Why Use Homebrew on macOS?

• Native macOS Experience: Integrates perfectly with your macOS workflow
• Simplified Installation: No more manual downloads or permissions issues
• Automatic Updates: Stay current with brew upgrade
• Clean Removal: Uninstall completely with brew uninstall

🚀 Getting Started on macOS

After installation:

1. Start the backend server with your language preference: meetily-server --language [your_language_code]
2. Launch Meetily from your Applications folder
3. Begin capturing and transcribing your meetings in your preferred language!

🔧 Advanced Configuration for macOS Users

Fine-tune your experience with additional parameters:

# Set logging level
meetily-server --language en --model medium --log-level debug
# Specify custom model path (if you've downloaded models elsewhere)
meetily-server --language en --model /path/to/your/model.bin

Note : This step installs the backend server and the frontend app.
Once the backend and the frontend are started, you can open the application from the Applications folder.

Option 2: Manual Installation

1. Download the dmg_darwin_arch64.zip zip file and extract to get meeting-minutes-frontend_0.1.0_aarch64.dmg file
2. Extract the file
3. Double-click meeting-minutes-frontend_0.1.0_aarch64.dmg file inside the extracted folder
4. Drag the meeting-minutes-frontend.app to Applications folder
5. Execute the following command in terminal to remove the quarantine attribute:

xattr -c /Applications/meeting-minutes-frontend.app

Backend

-- For Mac OS only

Brew Tap

brew tap zackriya-solutions/meetily

Brew install

brew install meetily-backend

Please add the anthropic or groq API key when asked.

To run the backend, first choose a model.
Available models: tiny, base, small, medium, large-v3

meetily-download-model medium

To run the backend

meetily-server

This should start the backend on Mac OS

You can set up API keys during installation or manually:

echo "ANTHROPIC_API_KEY=your_key_here" > $(brew --prefix)/opt/meetily-backend/backend/.env
echo "GROQ_API_KEY=your_key_here" >> $(brew --prefix)/opt/meetily-backend/backend/.env

Installing backend in Windows

1. Clone the repository:

git clone https://github.com/Zackriya-Solutions/meeting-minutes

2. Go to the backend directory:

3. Make sure you have installed all the prerequisites before proceeding:

   • ffmpeg
   • cmake
   • C++ compiler
   • Python versions between 3.10 and 3.12

   Windows-specific prerequisites:

   • Install Visual Studio Build Tools with C++ development workload
   • Install ffmpeg using Chocolatey: choco install ffmpeg
   • Ensure Python 3.10+ is in your PATH

4. Add environment file:

# On Windows (PowerShell):
"ANTHROPIC_API_KEY=your_api_key`nGROQ_API_KEY=your_api_key" | Out-File -FilePath .env -Encoding utf8

5. Add API Keys in .env files if you are planning to use Anthropic or Groq.

   • For Anthropic: Sign up at https://www.anthropic.com/ and get your API key
   • For Groq: Sign up at https://console.groq.com/ and get your API key from the API Keys section

6. Configure environment variables for Groq:

# On Windows (PowerShell):
$env:GROQ_API_KEY="your_groq_api_key"
# On Windows (Command Prompt):
set GROQ_API_KEY=your_groq_api_key

8. Build dependencies:

   • On Windows: .\build_whisper.bat (or run the commands from the script manually)

9. Run the server:

   • On Windows: .\start_with_output.ps1

What's Changed

• Windows audio capture support
• Cross-platform audio device detection improvements
• Enhanced error handling and logging
• Windows installer packages (.exe and .msi)
• Various bug fixes and stability improvements

Introducing subscription

We are planning to add a subscription option so that you don't have to run the backend on your own server. This will help you scale better and run the service 24/7. This is based on a few requests we received. If you are interested, please fill out the form here.

Full Changelog: v0.0.2...v0.0.3

What's Changed

• Optimize/main by @sujithatzackriya in #13
• BugFixes by @sujithatzackriya in #14
• Devtest by @sujithatzackriya in #15
• Update README with optimized demo GIF by @sujithatzackriya in #16
• Devtest to main by @sujithatzackriya in #17
• Hot fix: Removed old images and added new ones by @sujithatzackriya in #18
• Bug Fix : Install issues by @sujithatzackriya in #19
• Bug Fix: Fixed the microphone audio capture issue by @sujithatzackriya in #31
• Devtest to main after Optimize Audio Recording Player Performance by @sujithatzackriya in #40
• Windows OS Support addition Devtest update by @sujithatzackriya in #45

Full Changelog: v0.0.2...v0.0.3

Contributors

• @sujithatzackriya

Pre Release 0.0.2

A new release is available!

Last updated - March 1, 2025

What's New

• Transcription quality is improved.
• Bug fixes and improvements for frontend
• Better backend app build process
• Improved documentation
• New .dmg package
• New Windows packages for testing out the basic frontend

What would be next?

• Database connection to save meeting minutes
• Improve summarization quality for smaller llm models
• Add download options for meeting transcriptions
• Add download option for summary

Known issues

- Smaller LLMs can hallucinate, making summarization quality poor; Please use model above 32B parameter size

• Backend build process require CMake, C++ compiler, etc. Making it harder to build
• Backend build process require Python 3.10 or newer
• Frontend build process require Node.js

How to Get Started

Frontend

Running from dmg

1. Download the dmg.zip file
2. Extract the file
3. double click meeting-minutes-frontend_0.1.0_aarch64.dmg file inside the dmg folder
4. Move the meeting-minutes-frontend.app to Applications folder
5. Execute the following command in terminal

xattr -c /Applications/meeting-minutes-frontend.app

Backend

1. git clone the repo

git clone https://github.com/Zackriya-Solutions/meeting-minutes

2. Go to the backend directory:

cd meeting-minutes/backend
chmod +x build_whisper.sh 

3. Make sure you have install all the pre requsites before proceeding - ffmpeg, cmake, c++ compiler and Python versions between 3.10 and 3.12 are required

4. Add env file

echo -e "ANTHROPIC_API_KEY=your_api_key\nGROQ_API_KEY=your_api_key" | tee .env

4.1 Add API Keys in .env files if you are planning to use Anthropic or groq.

6. Build dependencies by eunning ./build_whisper.sh

7. Run the server with ./clean_start_backend.sh

What's Changed

• Pre release by @sujithatzackriya in #11

Full Changelog: v0.0.1.1...v0.0.2

Contributors

• @sujithatzackriya

Release v0.0.1 - Tauri Migration & First Stable Build

Overview

This release marks our transition from Electron to Tauri and delivers the first packaged executables. The changes focus on performance improvements and native integration capabilities.

Installation instructions

Install from source (Easier)

1. Get the Source code.zip file from the releases list

2. Extract the zip file

3. Open terminal and execute the following commands

Initial setup

cd meeting-minutes-0.0.1.1
# Give permission to run the whisper server
xattr -c ./backend/whisper-server-package/whisper-server 

Running backend

Pre requisites : Python 3.9 or 3.10

cd backend
# Create virtual environment
python3 -m venv venv
# Activate the venv
source venv/bin/activate 
# Install dependencies
pip install -r requirements.txt
# Add env file
echo -e "ANTHROPIC_API_KEY=your_api_key\nGROQ_API_KEY=your_api_key" | tee .env
# Run backend
./clean_start_backend.sh 
# Select whisper model and run the backend

Running from dmg

1. Download the dmg.zip file
2. Extract the file
3. double click meeting-minutes-frontend_0.1.0_aarch64.dmg file inside the dmg folder
4. Move the meeting-minutes-frontend.app to Applications folder
5. Execute the following command in terminal

xattr -c /Applications/meeting-minutes-frontend.app

6. Launch the application meeting-minutes-frontend from applications folder or by using spotlight search

Running frontend : Build from source

Pre requisites : npm, rust, pnpm

1. Open a new terminal
2. Execute following commands

 cd meeting-minutes-0.0.1.1/frontend 
 
 # Run frontend
 ./clean_build.sh 

Changelog

Frontend Architecture

• Migrated from Electron to Tauri
  • Native audio capture via Rust
  • 60% smaller installation size
  • System tray integration
  • Automatic updates
• Removed 600MB+ of Electron dependencies
• Added universal binary support (Apple Silicon/Intel)

Documentation Updates

• Updated README.md with:
  • New installation instructions
  • Tauri architecture diagram
  • Release notes section
  • Cross-platform build targets
• Revised docs/architecture.md to reflect new flow

Security

• Implemented process isolation
• Added memory-safe audio processing
• Removed Node.js dependency surface

Verification Steps

# Build verification
cargo check --release
cargo tauri build --target universal-apple-darwin
# Installation test
hdiutil verify meeting-minutes_0.0.1_universal.dmg
# Runtime checks
./src-tauri/target/release/meeting-minutes --validate
## What's Changed
* Updated for release by @sujithatzackriya in https://github.com/Zackriya-Solutions/meeting-minutes/pull/8
* Bug Fixes by @sujithatzackriya in https://github.com/Zackriya-Solutions/meeting-minutes/pull/9
**Full Changelog**: https://github.com/Zackriya-Solutions/meeting-minutes/compare/v0.0.1...v0.0.1.1