IrisFlow — An open source speech to text solution for…

Pitch

Description

Comments

About this project

IrisFlow is a powerful macOS speech-to-text overlay that enables seamless dictation across any app. With a simple Left-Ctrl command, users can speak and see their words transcribed in real-time. Designed for Apple Silicon and built for privacy, it processes data locally without the need for an account, ensuring a secure and efficient user experience.

IrisFlow: Open-Source macOS Speech-to-Text Overlay

IrisFlow is an innovative, open-source speech-to-text overlay specifically designed for macOS. This application enables users to seamlessly dictate text into any active application by holding the Left-Ctrl key, making it ideal for use in various contexts such as notes, emails, messages, and more.

Key Features

Local Dictation: No internet connection is needed. Users can speak freely and see their words transcribed in real-time into the focused application.
AI Mode: With AI mode activated, users can select text, hold Left-Ctrl, and request information vocally. The resulting answer can be effortlessly copied with a simple paste command (⌘V).
Custom Models: IrisFlow supports multiple AI models, including Cerebras, OpenAI, and Anthropic, allowing users to paste their API keys for a tailored experience.
Document Uploading: Users can upload PDFs and other documents, and with a valid Google AI key, they can search their content using AI mode.
Privacy-Centric: All user data, including chats and files, remain local to the Mac, with API keys securely encrypted using the macOS Keychain.

Compatibility

IrisFlow works on macOS 11 or later, specifically optimized for Apple Silicon (M1, M2, M3, M4). The application does not support Windows, Intel Macs, or Linux systems at this time.

Getting Started

To utilize IrisFlow, the following steps are recommended:

Install Node.js 18+ and required Command Line Tools on your Mac.

Clone the repository and build the application using the provided commands:

git clone https://github.com/joyboy5477/IrisFlow.git
cd IrisFlow/Irisflow
npm install
npm run install:mac

Open Iris Flow from the Applications folder, grant microphone access, and configure accessibility settings.

Typical Use Cases

IrisFlow is compatible with a range of applications, enabling users to dictate text into:

Note-taking apps: Notes, TextEdit, etc.
Web browsers: Chrome, Safari, Arc
Communication platforms: Slack, Discord
Code editors: VS Code, Cursor, Xcode

Conclusion

IrisFlow is a powerful tool for macOS users seeking an efficient and user-friendly dictation solution. By leveraging local processing and customizable AI integration, it streamlines the transcription process without compromising user privacy.