Aeterna: The Virtual Cemetery & Conversational Legacy Platform

🌟 Project Overview

Aeterna is a unique, next-generation virtual cemetery and memorial application that transforms static remembrance into dynamic, interactive conversation. Leveraging advanced conversational AI and voice cloning technology, Aeterna allows users to create persistent digital sanctuaries for loved ones or historical figures, enabling living relatives and the public to interact with their synthesized persona.

💡 Core Features

1. Geospatial Remembrance (The Map)

• Virtual Grave Mapping: Users can select a spot anywhere on a detailed interactive map (utilizing Mapbox/Google Maps API) and define a custom-shaped digital grave or memorial area.
• Profile Creation (Common Figures): Allows families to upload a common name, profile photo, life photos, and significant dates (e.g., birthday/death day).
• Privacy Controls: Critical toggle between Public (searchable by anyone) and Private (accessible only via a shared link or family circle).

2. The Legacy Core (Common Figures)

These features focus on building the rich, personal context needed to drive the conversational AI:

Feature	Description	Purpose
Life Photos	Dedicated gallery for organizing images and videos throughout the person’s life.	Visual documentation of the person’s narrative.
Context Upload	Ability to upload handwritten letters, diary entries, or documents. OCR is used to extract text and feed it to the LLM.	Injects personal syntax, specific memories, and philosophical viewpoints into the AI’s knowledge base.

3. Premium Conversational AI

The true innovation of Aeterna lies in its ability to bring a voice and personality back to the grave.

• Voice Cloning (ElevenLabs): Requires a minimum of 10 seconds of source audio (e.g., a voicemail or home video clip). This voice is cloned to synthesize all future conversation responses, complete with emotional cues like [thoughtfully], [sarcastically], or [whispers].
• Real-time Converse: Utilizing the ElevenLabs Conversational Agent, users can chat with the synthesized persona in real-time. The AI draws from the uploaded context and life narrative to provide meaningful, personalized answers.
• Read Loud Text: The persona can be prompted to read custom text, such as poems or recorded eulogies, in the deceased’s cloned voice.

4. Public Figure Profiles

• Curated Historical Personas (e.g., JFK): Dedicated, gold-standard profiles for notable historical figures.
• Interactive Dialogue: The public can converse with these personas, who are trained extensively on historical documents, speeches, and biographies to answer questions in character and in their distinctive speaking style (e.g., JFK’s accent).
• Public Access: These graves are permanently Public and serve as educational, interactive historical markers.

🛠️ Technical Stack (Hackathon Focus)

Component	Technology	Role
Voice Synthesis	ElevenLabs Voice Clone	Zero-shot, high-quality voice replication using minimal audio input.
Conversational Engine	ElevenLabs Conversational Agent	Provides low-latency, real-time response generation by coordinating STT, LLM, and TTS.
Geospatial Mapping	Mapbox GL JS / Google Maps API	Rendering 3D maps and handling custom polygon drawing for grave shapes.
Context Injection	Azure Computer Vision (OCR)	Extracts text from uploaded handwritten letters/documents to enrich the LLM prompt.
Database/Backend	Firebase / AWS Amplify	Hosting, authentication, and secure storage of sensitive media files.

✨ Vision & Impact

Aeterna aims to fundamentally shift the way we grieve and remember. By leveraging AI, the project offers a unique emotional and educational utility, ensuring that the voices, wisdom, and personalities of those we cherish remain accessible, creating a digital form of immortality.