CanopyRoute Pune — Offline Open-Source AI Trail Audio Runner

Описание и применение

CanopyRoute Pune — автономное приложение для трековых забегов, использующее локальные модели ИИ и процедурный звук, чтобы убрать экран из опыта и защитить личные данные.

Задача

Modern trail running apps keep runners glued to bright screens, exposing them to constant digital prompts and telemetry, which is problematic in areas with poor connectivity and raises privacy concerns.

Результат

The app delivers a fully offline, privacy‑preserving trail running experience that reduces screen distraction and works in areas without cellular coverage.

Рабочий процесс

• Select a trail and tap "Start Pocket Run"
• Slide phone into pocket; screen switches to low‑power OLED
• Browser runs local embeddings model (Xenova/all‑MiniLM‑L6‑v2) to rank trails based on natural language query
• Procedural audio engine synthesizes ambient sounds and speaks landmark cues via Web Audio and Web Speech APIs
• GPS milestones trigger audio cues and haptic feedback
• User runs without looking at the phone, relying on sound and vibration

Доказательства

CanopyRoute is engineered to run 100% locally on the runner's device with zero external server dependencies

Источник

Runs Xenova/all-MiniLM-L6-v2 locally in the browser using WebAssembly / WebGPU via ONNX Runtime Web

Источник

Instead of streaming pre-recorded audio files over cellular networks, CanopyRoute uses the Web Audio API to procedurally synthesize ambient mountain breezes and wind noise

Источник

With closed cloud APIs, this data is sent over the wire to remote servers. With open-source on-device AI, zero bytes of biometric or location data ever leave the runner's phone

Источник