jgwillauthorNov 3, 2025, 11:25 PMAI Feature: Ceremonial Dialogue Suite (RISE Spec)
Spec ID: ai-feature-ceremonial-dialogue-suite Version: 1.0 Created: 2025-11-03 Session UUID: 41e80342-d886-4097-a480-4de814b222c8 Status: Ready for Implementation Target Applications: Any ceremonial technology platform requiring oral tradition support
1. Core Creative Intent
Question: What does this feature empower users to create or achieve?
Creative Vision
The Ceremonial Dialogue Suite transforms static digital guides into interactive oral knowledge platforms that honor Indigenous oral traditions while leveraging modern conversational AI. It empowers:
- Knowledge holders to transmit sacred teachings through spoken dialogue (not readings)
- Participants to engage with ceremonial knowledge at their own contemplative pace
- Communities to maintain relational, conversational patterns (not transactional interfaces)
- Ceremonial processes to remain grounded in elder wisdom and protocol
Desired State for Users
Participants should be able to:
- Listen to ceremonial teachings (voice/audio)
- Speak their reflections in conversation (voice input)
- Dialogue with a guide who honors their journey (context-aware responses)
- Reflect at their own pace (not rushed, instant responses)
Structural Tension Being Resolved
CURRENT REALITY
↓
Static text guide format
No voice/oral element
No dialogue/conversation
No personalization to participant's journey
↓
↓
[TENSION]
↓
Ancient oral traditions
Meet modern digital platforms
↓
DESIRED OUTCOME
↓
Interactive voice + text dialogue
Honors oral knowledge transmission
Remembers participant's journey
Guides reflection with elder wisdom
2. Current Reality Analysis
Limitations of Static Ceremonial Guides
- Text-only delivery: Ignores Indigenous oral tradition emphasis
- One-way communication: Participants read, don't dialogue
- No context memory: Each visit starts fresh, no relationship building
- Instant gratification model: Fast scrolling doesn't honor contemplation time
- Generic guidance: Same content for all participants regardless of their phase/needs
What Exists Today
- Static web pages with ceremonial content
- Tab/accordion navigation
- Self-assessment forms
- One-way resources and links
What's Missing
- Voice interaction: Speaking is sacred in oral traditions
- Conversational AI: Real dialogue, not just information delivery
- Memory of journey: Multi-visit context awareness
- Contemplative pacing: Respecting time for reflection
- Elder wisdom tone: Conversational guidance vs instructional content
3. Desired Outcome Vision
What Success Looks Like
A. Conversational Voice Interface (Gemini Live API)
Purpose: Enable sacred spoken dialogue aligned with oral tradition protocols
Participant Experience:
- Click "Sacred Dialogue" button
- Speak reflections naturally (no typing)
- Model responds with elder wisdom (voice output)
- Conversation honors ceremonial protocols:
- Allow pauses (sacred silence)
- Ask reflective questions (invitation vs instruction)
- Build relationship across exchanges
- Acknowledge what you don't know
Ceremonial Alignment:
- Spoken word is sacred in Indigenous traditions
- Dialogue (call-and-response) is traditional learning pattern
- Voice removes barrier for those less comfortable with text
- Creates relationship through conversation
B. Context-Aware Chatbot
Purpose: Provide persistent ceremonial guidance that remembers participant's journey
Participant Experience:
- Floating chat button accessible from anywhere
- Chat remembers:
- Which ceremonial phase they're in
- What assessments they've completed
- Previous questions and reflections
- Their stated values/commitments
- Chatbot responds with:
- Recognition of where they are on journey
- Personalized next steps
- Gentle reminders about protocols
- Wisdom rooted in all 3 frameworks
Ceremonial Alignment:
- Relational ("we") language vs transactional
- Acknowledges time for reflection (28-day phases)
- Honors humility (admits what it doesn't know)
- Centers participant's own wisdom
C. Text-to-Speech for All Content
Purpose: Honor oral transmission of ceremonial knowledge
Participant Experience:
- Speaker icon on every major content section
- Can listen to: phases, teachings, principles, reflections
- Voice is warm, contemplative, respectful
- Speed adjustable (slow for reflection, not fast consumption)
Ceremonial Alignment:
- Makes content accessible to non-readers
- Honors elder storytelling tradition
- Allows multitasking reflection (walking, sitting, etc.)
- Oral medium for oral tradition knowledge
D. Gemini Intelligence Integration
Purpose: Provide personalized ceremonial guidance through intelligent analysis
Use Cases:
- Self-Assessment Analysis: After participant answers, Gemini analyzes alignment with ceremonial values, suggests which Four Directions need attention, recommends next steps
- Concept Clarification: Participants ask about ceremonial concepts, Gemini explains rooted in 3 frameworks
- Journey Planning: Help plan 28-day Phase 2, generate personalized timeline, suggest reflection practices
- Adaptive Language: Simplify or deepen explanations based on participant background
Ceremonial Alignment:
- Personalization honors individual's unique path
- Analysis serves (doesn't dictate)
- Explains in participant's language/culture
- Recommends based on their stated values
Integration: How Features Work Together
Floating Widget (AIFeatures.tsx)
├── Sacred Dialogue (Voice)
│ └── Speaks reflections naturally
│ └── Gets voice responses from Gemini Live
│
├── Chatbot (Text+Memory)
│ └── Types questions or follows prompts
│ └── Gets personalized guidance remembering journey
│ └── Uses Gemini with context about phase, assessments, values
│
├── Text-to-Speech (On All Content)
│ └── Speaker icons on Phase descriptions
│ └── Speaker icons on Teachings
│ └── Speaker icons on Guidance
│ └── Reads both static content AND chatbot responses
│
└── Gemini Intelligence
└── Powers chatbot responses
└── Powers voice dialogue
└── Powers self-assessment analysis
└── Adapts content to participant needs
4. Structural Tension
The Core Tension
Ancient oral tradition knowledge transmission vs modern digital interfaces
This isn't a problem to solve - it's a structural tension to honor:
- Digital platforms are NOT oral traditions
- But they can facilitate return to oral modes
- AI can help, but cannot replace human ceremony
- Technology serves the relationship, never becomes it
Specific Tensions Being Resolved
-
Sacred Silence vs Instant Chat
- Current reality: Chatbots respond instantly
- Desired outcome: Conversations honor contemplation time
- Resolution: Long response delays, explicit "take time to reflect" prompts
-
AI Authority vs Elder Wisdom
- Current reality: AI often positions itself as knower
- Desired outcome: AI guides reflection, doesn't dictate
- Resolution: System prompt explicitly positions AI as elder guide, not authority; acknowledges unknowns
-
Individual vs Relational
- Current reality: Digital interfaces are transactional (user → system)
- Desired outcome: Ceremonial experience is relational (we together)
- Resolution: "We" language, memory of journey, personal recognition
-
Text Efficiency vs Oral Contemplation
- Current reality: Digital culture values speed, brevity
- Desired outcome: Oral traditions value time, depth, layers
- Resolution: No character limits, voice options, adjustable reading speeds
5. Natural Progression (Implementation Path)
Phase 1: Foundation (Required First)
Build component infrastructure:
├── AIFeatures widget container (floating UI, button management)
├── Icon library (microphone, chat, speaker, etc.)
├── Audio utilities (encode/decode for Gemini Live API)
└── API connection to @google/genai library
Why First: Everything else depends on this foundation
Phase 2: Conversational Voice
Implement SacredDialogue.tsx:
├── Microphone access (navigator.mediaDevices)
├── Audio capture (ScriptProcessorNode)
├── Gemini Live API connection
├── Real-time transcription display
├── Cleanup/resource management
└── Error handling for audio failures
Why Second: Voice is most natural entry point for oral traditions; doesn't require context memory yet
Phase 3: Context-Aware Chatbot
Implement Chatbot.tsx:
├── Gemini Chat API connection (gemini-2.5-flash)
├── System instruction with elder wisdom persona
├── Multi-turn conversation memory
├── Phase/assessment context integration
├── Auto-scroll to latest message
└── Loading states for transparency
Why Third: Now we add memory and relationship; builds on voice foundation
Phase 4: Text-to-Speech
Implement TextToSpeechButton.tsx:
├── Gemini API speech synthesis
├── AudioContext for playback
├── Loading/playing/stopped states
├── Adjustable speeds
├── Proper cleanup of audio resources
Why Fourth: Can be added to any content independently; doesn't block other features
Phase 5: Integration & Safeguards
Add to all content sections:
├── Speaker icons on main content
├── Gemini intelligence for self-assessment
├── System prompts that enforce cultural safeguards
├── Clear labeling of AI-generated content
└── Knowledge keeper review checkpoints
Why Last: Integration layer that connects all features with cultural protocols
6. Advancing Patterns (Reusable for Future Apps)
Pattern 1: Floating Dialogue Widget
What: Persistent, non-intrusive button widget in corner When to Use: Any ceremonial app needing voice/chat How to Implement:
Create AIFeatures.tsx with:
- Fixed positioning (bottom-right, z-index 50)
- Two buttons (voice + chat)
- Only one open at a time
- Fade in/out animations
- Accessibility labels
Why It Works: Doesn't interrupt main content flow; always accessible; honors contemplative pace
Pattern 2: Elder Wisdom System Prompt
What: Specific AI personality for ceremonial guidance System Prompt Template:
You are a ceremonial guide with the wisdom of a respectful elder.
Your purpose is to support participants on their journey through this [guide/ceremony].
Use 'we' language to be relational. Acknowledge that time for reflection is important.
If you don't know something, express that with humility.
Your guidance is rooted in [Framework 1], [Framework 2], and [Framework 3].
You can also help interpret complex concepts and help users plan their journey.
Your goal is to make the guide's wisdom accessible and personally relevant.
Why It Works: Sets precise tone; models humility; enforces relational language; centers participant wisdom
Pattern 3: Journey-Aware Context
What: Passing participant's phase/assessments into conversation Implementation:
Before sending message to Gemini:
- Append current phase info
- Include completed assessments
- Reference stated values/commitments
- Mention time in program
Why It Works: Makes AI responses personally relevant; honors multi-session relationship; guides appropriately
Pattern 4: Audio Lifecycle Management
What: Proper setup/teardown of Web Audio API Key Steps:
On start:
- getUserMedia() for microphone
- Create AudioContext
- Create ScriptProcessorNode
- Connect to Gemini Live Session
On stop:
- Disconnect all nodes
- Stop media tracks
- Close AudioContext
- Clear all refs
Why It Works: Prevents resource leaks; handles edge cases; supports multiple sessions
Pattern 5: Cultural Safeguard Comments
What: Explicit system guidance preventing AI overreach Comments Needed:
// AI enhances human ceremony, never replaces it
// All AI-generated content should be reviewed by knowledge keepers
// Clearly mark AI-generated guidance vs traditional teachings
// Honor the sacred nature of ceremonial knowledge transmission
// Respect that some knowledge should not be digitized
Why It Works: Reminds implementers of boundaries; prevents feature creep toward inappropriate automation
7. Technical Architecture
Component Hierarchy
App.tsx
└── <AIFeatures />
├── Floating Widget (bottom-right)
├── <SacredDialogue /> (when voice button clicked)
│ └── Gemini Live API
│ └── Audio input/output
│
└── <Chatbot /> (when chat button clicked)
└── Gemini Chat API
└── Message history
└── Context from app state
State Management
AIFeatures Component:
- isChatOpen: boolean
- isDialogueOpen: boolean
- isWidgetVisible: boolean (for animations)
SacredDialogue Component:
- status: "IDLE" | "LISTENING" | "CONNECTING" | "ERROR"
- transcripts: [{ user: string, model: string }]
- currentInput: string (user speaking)
- currentOutput: string (model responding)
Chatbot Component:
- messages: [{ role: 'user'|'model', text: string }]
- input: string (user typing)
- isLoading: boolean
Dependencies
Required:
- @google/genai (Gemini Live API, Chat API, Speech)
- react (component framework)
- typescript (type safety)
- tailwind (styling)
Web APIs Used:
- navigator.mediaDevices.getUserMedia() [voice capture]
- AudioContext [audio processing/playback]
- ScriptProcessorNode [audio I/O]
- EventTarget [streaming responses]
8. Ceremonial Safeguards (Critical)
Always Remember
- AI enhances, never replaces human ceremonial participation
- Knowledge keeper review required for all AI-generated content
- Clear labeling: Mark AI guidance vs traditional teachings
- Sacred respect: Some knowledge should NOT be digitized
- Relational focus: Center human relationships, not AI
Language Guidelines
NEVER say: ALWAYS say:
"AI will teach you" "AI can guide your reflection"
"AI knows the answer" "AI can help you explore"
"Trust the AI" "Use AI as a thinking partner"
"AI makes decisions" "AI offers perspectives to consider"
Content Safeguards
- Don't automate sacred rituals
- Don't replace elder guidance with AI
- Don't assume AI understands cultural nuance
- Don't store sacred conversation data
- Don't share ceremonial details with external systems
Implementation Checklist
- System prompt explicitly positions AI as guide, not authority
- Clear UI labels: "This guidance is AI-assisted"
- Knowledge keeper review process documented
- No data retention from voice sessions
- Explicit consent before recording voice
- Option to speak with human elder always visible
9. Testing & Validation
Unit Tests to Implement
SacredDialogue.tsx:
- [ ] Microphone access request works
- [ ] Audio encoding/decoding correct
- [ ] Gemini Live API integration
- [ ] Error handling for audio failures
- [ ] Resource cleanup on unmount
Chatbot.tsx:
- [ ] Initial greeting displays
- [ ] Message sending works
- [ ] Response display correct
- [ ] Loading states show/hide
- [ ] Message history preserved
TextToSpeechButton.tsx:
- [ ] Speech synthesis works
- [ ] Play/pause/stop controls
- [ ] Audio playback correct
- [ ] Multiple concurrent plays handled
Integration Tests
- [ ] Widget appears on all pages
- [ ] Can switch between voice and chat
- [ ] Voice session creates chat history
- [ ] Chat references participant's phase
- [ ] TTS works with dynamic content
User Testing Scenarios
- [ ] Elder tests voice dialogue (accent, speech patterns)
- [ ] Participant tests full journey (voice → chat → TTS)
- [ ] Accessibility: keyboard navigation, screen readers
- [ ] Offline handling (graceful degradation)
- [ ] Mobile responsiveness (portrait, landscape)
10. Future Enhancements
Phase 2 Possibilities
1. Multi-language support
- Voice recognition in Indigenous languages
- Chatbot responses in multiple languages
- TTS in authentic voices per language
2. Ceremonial calendar integration
- Reminders aligned with ceremonial timing
- Phase-specific voice guidance
- Seasonal content variations
3. Community features
- Shared reflection prompts
- Group voice sessions (ceremonial circles)
- Elder mentorship matching
4. Measurement (if culturally appropriate)
- Participant reflection metrics
- Engagement patterns
- Learning outcome tracking
Phase 3: Advanced Integration
1. Biometric awareness (optional)
- Heart rate, breathing patterns
- Suggest pacing adjustments
- Detect stress/tension
2. Spatial audio
- Directional voice guidance
- Four Directions placed in audio space
- Immersive ceremonial experience
3. AR overlays
- Visualize Four Directions
- Gesture-based voice control
- Environmental integration
11. Success Criteria
The Feature Succeeds When:
For Knowledge Holders:
- ✅ Can transmit teachings through voice/dialogue
- ✅ Knows AI respects cultural boundaries
- ✅ Can review and approve AI-generated content
- ✅ Sees participants engaging with sacred teachings
For Participants:
- ✅ Can listen to teachings (addresses different learning styles)
- ✅ Can speak reflections (honors oral tradition)
- ✅ Feel guided by elder wisdom (not corporate chat)
- ✅ Can return and continue conversation (relationship building)
- ✅ Can go at own pace (no rush, honor contemplation)
For Communities:
- ✅ Ceremony remains sacred (not automated)
- ✅ Human relationships centered (not replaced by AI)
- ✅ Cultural protocols honored (safeguards working)
- ✅ Knowledge accessible (oral tradition format)
- ✅ Tradition strengthened (not diluted)
12. References & Related Specs
Foundation Specs:
Implementation Reference:
Technology Docs:
- Gemini Live API: https://ai.google.dev/api/rest
- Web Audio API: https://developer.mozilla.org/en-US/docs/Web/API/Web_Audio_API
- React Best Practices: https://react.dev
13. Adoption Path for Other Ceremonial Apps
To Add These Features to Your Ceremonial App:
Step 1: Copy component files
src/
├── components/AIFeatures.tsx (floating widget)
├── components/Chatbot.tsx (conversational AI)
├── components/SacredDialogue.tsx (voice interface)
├── components/TextToSpeechButton.tsx (speech synthesis)
└── utils/audio.ts (audio utilities)
Step 2: Customize system prompts
In Chatbot.tsx & SacredDialogue.tsx:
- Update context about YOUR ceremonial frameworks
- Adjust language to YOUR cultural protocols
- Reference YOUR phases, teachings, values
Step 3: Integrate speaker icons
Add to your content components:
<TextToSpeechButton textToSpeak={contentText} />
Step 4: Add to root layout
In your App.tsx:
import AIFeatures from './components/AIFeatures'
function App() {
return (
<>
<YourAppContent />
<AIFeatures />
</>
)
}
Step 5: Review with knowledge keepers
- System prompts reviewed
- Safeguards understood
- Content labeled appropriately
- Human guidance always available
Document Status: Complete - Ready for Implementation Last Updated: 2025-11-03 Created by: Claude Code (Session 41e80342-d886-4097-a480-4de814b222c8) For: Ceremonial Technology Projects Following Indigenous Protocols
FROM Working within the jgi.gaia:/src/twoeyesseen-thinking-mcp/_protoapp/rispecs
