DeafBlind Video Assistant 1.0.0

The DeafBlind Video Assistant is a powerful desktop application designed to make video content more accessible to DeafBlind individuals. This release marks our first stable version, offering a comprehensive suite of features for video processing and AI-powered content analysis.

Key Features

1. Video Processing

Seamlessly merge subtitle files (SRT, ASS) with videos
Hardware-accelerated video processing using FFmpeg
Optimized subtitle positioning and visibility
Optional black bars for enhanced subtitle readability
Real-time progress tracking during processing

2. AI-Powered Visual Description

Advanced visual analysis using Google's Gemini AI
Detailed scene descriptions focusing on:
- Visual setup and environment
- People present and their positioning
- Relevant visual context and background details
Chronological narrative combining visual and subtitle content
Customizable description prompts

3. User Experience

Intuitive, accessible interface
Clear task selection and workflow
Real-time progress feedback
Secure API key management
Persistent settings storage
Copy-to-clipboard functionality
High-contrast visual design

System Requirements

macOS 10.13 or later
Apple Silicon (M1/M2/M3) processor
Internet connection for AI features
Google AI Studio API key for visual description features
Minimum 4GB RAM recommended
500MB free disk space

Installation Instructions

Download the DMG file
Open the DMG file
Drag "DeafBlind Video Assistant" to your Applications folder
Launch from Applications folder
On first launch:
- Allow any required system permissions
- Visit Settings to configure your Gemini API key
- Test API key connection before use

Getting Started

Obtain a Gemini API key from Google AI Studio
Enter your API key in the Settings
Choose your task:
- "Generate Transcription" for AI-powered video descriptions
- "Merge Subtitles" for embedding subtitles into videos
Follow the on-screen instructions for file selection and processing

Security & Privacy

Application is signed with Apple Developer ID
Local processing of video files
Secure API key storage
No data collection or tracking

Support & Feedback

For issues, feature requests, or feedback, please create an issue in the repository.

Acknowledgments

Special thanks to the DeafBlind community for their invaluable input in shaping this tool's development and ensuring its effectiveness in real-world scenarios.

SHA256: b5158bdf30a5de2cee08b2317d14af155fb39f9a5f416ee44d3736b45692cfde
Build Date: January 28, 2025

2.6 KiB Raw Permalink Blame History