Trailtale: A Pocket-Sized Audio Story That Gets You Off Your Screen and Back on the Trail
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built Trailtale is a local, offline walking app that turns a walk into an immersive audio story. While you move, the app plays a short story segment; when you stop, the audio pauses automatically. It also surfaces weekly nature quests and lets you confirm interesting finds with a photo, all without…
Trailtale is an offline walking app that transforms a stroll into an engaging audio narrative. While walking, the app delivers distinct story segments; upon pausing, the audio halts automatically. Weekly nature quests are presented, and users can document discoveries with photos, all without needing to return to a screen. The concept revolves around immersing users in their walk experience without distractions from their phone.
The project is available on the GitHub repository at HarshitAnand1 / trailtale-touch-grass. Trailtale operates solely on the device, utilizing local models and no cloud infrastructure, catering to privacy, local control, and adaptability requirements. The app avoids network calls for text-to-speech (TTS) and photo verification, minimizing costs and preserving user data.
Users can opt for llama.cpp for story generation, or Piper for TTS, both of which allow for customization. The architecture comprises a PWA client on the phone, coupled with a FastAPI backend server managing the walk flow, story queue, quests, and logging. Story generation relies on a local model through llama.cpp, while voice generation is handled by Piper, with fallback text provided in case of model unavailability.
The quest system features weekly nature quests stored as local JSON files, tracked in an SQLite database. An optional multimodal model can confirm whether a photo aligns with a quest item, all while ensuring the app functions offline.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.