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
