SmritiPath: Turning Family Memories of a Place into a Walk, with Local AI
Description and use
SmritiPath turns family memories of a place into a walking route using local open‑source models, preserving privacy and enabling offline use.
Problem
People want to revisit how a place used to look by walking with someone who remembers it, but existing tools require cloud AI services and data upload.
Outcome
SmritiPath enables users to create a walking route based on personal memories without sending data to external services, preserving privacy and reducing cost.
Workflow
- Add a recording (faster-whisper) or paste a memory as text
- A local model (gemma3:4b via Ollama) finds each place mentioned and extracts a quote
- Pin the stops on a map (Leaflet + OpenStreetMap) and reorder the route
- Download the Walk Pack, an offline HTML page with a sketch map, stories, questions, and blank "Now" lines
Evidence
Speech-to-text and stop-finding run locally with open-source models. Nothing is sent to any AI service.
A local model (gemma3:4b via Ollama) finds each place mentioned, with a quote from the speaker. Code checks every quote really appears in the transcript.
- Record date: 2026-10-10T22:41:58.833Z