Inspiration

The creator economy is booming with over 50 million content creators worldwide, but most struggle with the same problem: creating engaging, SEO optimized YouTube content which takes hours of research, writing, and optimization. We watched talented creators spend more time on content planning than actual video creation. So I built KeyTube which transforms any web content into professional YouTube videos!

What it does

KeyTube transforms any web content into professional YouTube ready material in seconds. Users can highlight text on any webpage or speak their ideas, and our extension generates catchy titles, engaging scripts, SEO-optimized descriptions, and relevant tags. It supports 5 languages with real time translation, offers speech to text input for creators on the go, and exports everything to professional PDFs. Whether you're researching competitors, reading news articles, or browsing educational content, KeyTube turns that inspiration into actionable video content instantly.

How we built it

We built KeyTube as a Chrome Manifest V3 extension using modern JavaScript and Chrome's cutting-edge built-in AI APIs. The core architecture leverages the Prompt API (Gemini Nano) for content generation and the Translator API for multi language support. We implemented a dual mode system: Research Mode extracts content from webpages using content scripts, while Manual Mode accepts direct input with Web Speech API integration for voice transcription. The UI uses Chrome's SidePanel API for a modern, responsive experience with custom CSS animations and theme support. We integrated jsPDF for professional document export and implemented sophisticated error handling with graceful fallbacks.

Challenges we ran into

  • Content Extraction: Different websites structure content differently, so we developed a sophisticated content extraction algorithm that works across news sites, blogs, and educational platforms while filtering out UI elements and advertisements.
  • Real-time Performance: Balancing AI processing speed with user experience required implementing streaming text generation and smart caching to ensure responsive interactions.
  • The most challenging part was creating a professional pdf export.

Accomplishments that we're proud of

  • Seamless AI Integration: Successfully implemented both Prompt and Translator APIs with streaming generation and real time feedback
  • Professional User Experience: Created a polished and responsive interface
  • Multimodal Input: Combined text, voice, and web content extraction into a unified workflow
  • Global Accessibility: Built-in translation support for 5 languages makes content creation accessible worldwide
  • Privacy First Design: 100% local processing means user data never leaves their device
  • Production Ready: Comprehensive error handling, accessibility features, and professional documentation
  • Real world Impact: Reduces content creation time from hours to minutes while improving quality

What we learned

  • Chrome's AI APIs are amazing: The power of local AI processing is incredible fast, private, and always available. The Prompt API's streaming capabilities enable real time user experiences that feel magical.
  • User Experience is Everything: Even the most powerful AI is useless if users can't access it easily. We learned the importance of clear onboarding, helpful error messages, and intuitive workflows.
  • Content Creation is Complex: YouTube optimization involves understanding SEO, audience psychology, and platform algorithms. Our AI prompts needed extensive refinement to generate truly effective content.
  • Privacy Matters: Users are increasingly concerned about data privacy. Chrome's built-in AI allows us to offer powerful features without compromising user data.
  • Accessibility is Essential: Building for diverse users with different abilities and languages taught us to think inclusively from the start.

What's next for KeyTube

  • Enhanced AI Capabilities: As Chrome's AI APIs evolve, we'll integrate image analysis for visual content inspiration and advanced multimodal features.
  • Platform Expansion: Extend beyond YouTube to support TikTok, Instagram, LinkedIn, and other social platforms.
  • Advanced Personalization: Learn from user preferences to generate increasingly personalized content suggestions.

APIs Used

  1. Prompt API: Core content generation using Gemini Nano
  2. Translator API: Real-time multi-language translation

Supporting Technologies

  1. Web Speech API : For real time speech to text transcription
  2. SidePanel API : For modern UI

Made with ❤️ for content creators using Chrome's Built-in AI

Built With

Share this project:

Updates