/lifescope-voice

Record and gain insights with your voice

Primary LanguageJavaScriptMIT LicenseMIT

LifeScope-VOICE

Repository

(development phase, low priority)

The LifeScope API allows anyone to create a shared virtual space to access their memory and tell their stories.

This repo contains a series of voice command search applets written to extend the lifescope-app codebase as JS/Vue/Nuxt plugins.

Browser Web Speech API allows for speech recognition and synthesis support in Chrome, Edge, Firefox, Safari (partial). This API is super badass but no one ever uses it.

Requirements

  • MVP: Respond to simple voice commands

    • Examples
      • "Run Saved Search: [Search Name]"
      • "Show me [View Name]"
      • "Add [Provider Name]"
      • "What was the last thing I did?"
      • "What did I do today?"
      • "Tell me about [Person Name] and I"
    • Voice Response
      • "28 Events match your saved search"
      • "Today's List of events. You received a message from Jane at 5:44 pm saying...."

  • Deep Voice Search

    • MVP: Structured search terms (who, what, when, where, why)
    • Natural Language Processing.

  • Passively capture voice data (speech text and recording) and store in LifeScope.

  • AI Storytelling.

    • "Here were the highlights from your trip to Hawaii...."

Dependencies

  • Vue/Nuxt compatible

Examples