Skip to content
Ayush08kPublic

About

VibeWalk a digital step counter application with AI integration for Multiple Sections

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

Β 

History

15 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ‘Ÿ VibeWalk β€” Next-Gen AI Step Tracker & Health Analytics

VibeWalk is a modern, high-performance React Native mobile application built on Expo SDK 54 paired with an AI Analytics Engine. It tracks daily steps, calories burned, distance covered, active duration, walking cadence, GPS routes, achievement milestones, and 30-day activity trends with a futuristic Electric Cyan & Cyber Violet OLED dark interface, smooth liquid flow UI-thread transitions, audio coaching & cadence soundscapes, and hardware sensor step filtering.


🌟 Key Features

1. 🎯 Holographic Concentric Dual-Ring Hero Dial

  • Outer Progress Ring: Visual step goal arc (#00F5FF β†’ #9D00FF β†’ #FF007A) powered by react-native-reanimated spring physics.
  • Inner Sub-Arc: Sub-metric calorie target track.
  • Center Readout: Ultra-bold hero step count, target goal readout (TARGET 10,000), mini progress bar, and dynamic status tags (⚑ GOAL CRUSHED, πŸ”₯ ALMOST THERE).

2. 🌊 Liquid Flow Transitions & Staggered Section Animations

  • Tab Focus Morphing (LiquidScreenWrapper.tsx): Fluid UI-thread slide, scale, and cross-fade animations on tab switches powered by react-native-reanimated and @react-navigation/native.
  • Staggered Wave Sections (LiquidSection.tsx): Cards, grids, and timeline elements enter with organic staggered spring physics (FadeInDown.springify().damping(13)) and touch rebound interaction.

3. 🎡 Cadence Beat Sync & Audio Voice Cues

  • Live Audio Coach (voiceCoachService.ts): Speech-synthesized text-to-speech updates using expo-speech announcing session start, pause, finish, and per-kilometer split updates ("Kilometer 1 complete. Average pace: 5'32" per kilometer").
  • Cadence Soundscapes (soundscapeService.ts): Ambient audio tracks (Synthwave Pulse, Cyber Rain, Lo-Fi Walk Groove) featuring a dynamic tempo multiplier (0.85x to 1.35x) adjusting live to your walking cadence (steps/min).

4. πŸ—ΊοΈ Live GPS Route Tracker & GPX Export

  • Real-Time Mapping: Interactive map integration displaying real-time user movement polyline trails (#00F5FF).
  • GPX Export (gpxService.ts): Export tracked workout sessions into standard .gpx XML files ready for Strava, Garmin, or Apple Health import using expo-sharing.
  • Bookmarked Favorite Trails (favoriteTrailsService.ts): Bookmark AI-suggested routes in the AI Planner (⭐ Save Trail) and select them on the GPS Walk screen via a horizontal trail carousel.

5. 🌀️ Weather & Air Quality Telemetry

  • Environmental Status Card (weatherService.ts): Displays live outdoor environmental conditions on the GPS Walk screen including Temperature (Β°C/Β°F), Weather Condition, Humidity %, UV Index rating, Air Quality Index (AQI 1-5), and outdoor walking safety advice.

6. πŸ“Š 4-Grid Activity Metrics & Action Dock

  • πŸ”₯ Calories: Burned kilocalories with individual progress mini-fill (#FF007A).
  • πŸ“ Distance: Kilometers covered with pace calculations (#00F5FF).
  • ⏱ Active Duration: Active walking minutes (#9D00FF).
  • ⚑ Walk Pace: Real-time walking cadence in steps / min (#FF9900).
  • Action Control Dock: Floating pill dock featuring +100 Quick Steps and Auto Walk mode toggle for instant testing on desks or emulators.

7. πŸ“… Interactive 7-Day Week Selector & 30-Day Activity Chart

  • Interactive Week Selector: Tap any weekday pill (MON – SUN) in History to inspect that specific day's step breakdown, date stamp, and target achievement ratio.
  • Dual-Gradient 30-Day Chart: Built with react-native-gifted-charts, featuring a target reference threshold line, average steps pill badge, and color-coded bars (Cyan for goal met, Violet for moderate, Red/Orange for low).
  • Weekly Pace Comparison Split Card: Dual progress bars comparing this week's daily average versus last week's daily average with trend percentage diff badges (+14% β†—).

8. πŸ† Gamified Achievement & Badge System

  • Tiered Milestone Badges: Unlockable achievements categorized across step goals, streak milestones, distance benchmarks, and active time consistency.
  • Visual Progress Tiers: Color-coded badges (#00F5FF, #9D00FF, #FF007A) showing completion percentage, unlocked dates, and locked criteria previews.

9. πŸ€– AI Health Planner & Goal Recommendations

  • Dynamic Weekly Schedules: Customized daily step targets based on recent activity performance and AI wellness score recommendations.
  • Route Suggestions & Window Times: Recommends optimal circadian walking time windows and neighborhood routes with one-tap trail bookmarking.

10. πŸ“² Lock Screen Live Activities & Dynamic Island Controller

  • System Widget Previews: Interactive preview component (WidgetPreviewCard.tsx) rendering Home Screen, Lock Screen, and Dynamic Island ongoing notification controls.
  • Quick Metrics Glance: Displays live step count, ring completion arc, and active timer telemetry directly in widget formats.

11. πŸ›‘οΈ Rhythmic Cadence Step Filter (Zero False Steps While Sitting)

  • Impact Acceleration Threshold (1.42g): Prevents slight hand movements or wrist turns while sitting/resting from triggering false steps.
  • Rhythmic Step Cadence Window ([320ms, 1100ms]): Requires at least 3 consecutive rhythmic steps matching human walking cadence before steps are counted. Isolated hand twitches while resting or sleeping are completely discarded.

πŸ€– Where AI Is Used in VibeWalk

AI power is integrated at both the Backend Engine and Client-Side Fallback System:

+-------------------------------------------------------------------+
|                        VibeWalk App Flow                          |
+-------------------------------------------------------------------+
                                  |
                   [30-Day Activity Vector Data]
                                  |
                                  v
+-------------------------------------------------------------------+
|                      AI Analytics Engine                          |
|             (FastAPI / Express or Offline AI Fallback)            |
+-------------------------------------------------------------------+
    |                             |                            |
    v                             v                            v
[Pattern Analysis &      [0-100 Dynamic Wellness     [Ranked Contextual AI
 Trend Mining]            Performance Score]          Insight Generation]
    |                             |                            |
    +-----------------------------+----------------------------+
                                  |
                                  v
+-------------------------------------------------------------------+
|                      Cyber Frontend Timeline                      |
|       (Renders Ranked AI Insights: Success, Warning, Alert)       |
+-------------------------------------------------------------------+

1. πŸ“ˆ Pattern Analysis & Trend Mining

  • The AI engine processes 30-day historical step arrays [s1, s2, ..., s30] sent by the mobile app.
  • Calculates exponential moving averages (EMA), rolling 7-day averages, and week-over-week variance to detect upward momentum (up), activity dips (down), or steady consistency (stable).

2. πŸ† Dynamic 0–100 Wellness Performance Scoring

  • Evaluates four weighted dimensions:
    • Goal Consistency Rate (weight: 50%)
    • Active Streak Multiplier (weight: 20%)
    • Week-over-Week Growth Rate (weight: 15%)
    • Volume Stability (weight: 15%)
  • Assigns dynamic performance tier badges: OPTIMAL (85–100), HEALTHY (65–84), MODERATE (45–64), and LOW ACTIVITY (0–44).

3. πŸ’‘ Contextual Ranked AI Health Insights

  • Generates severity-ranked health cards with emoji tags and actionable coaching:
    • success (πŸ”₯ Streak & πŸš€ Momentum): "You've hit your 10,000 step goal 5 days in a row!"
    • warning (⚠️ Activity Dip): "Your steps dropped 18% compared to last week."
    • alert (🚨 Inactivity Dip): Detects multi-day inactivity and suggests recovery targets.
    • info (πŸ“Š Benchmark Insights): Compares weekend vs. weekday walking patterns.

4. ⚑ Embedded Client-Side Offline AI Engine

  • If the remote AI backend is unreachable (offline mode or dev network change), apiService.ts automatically executes an embedded client-side AI analysis algorithm so users always receive accurate wellness scores, trends, and smart insights.

πŸ› οΈ Tech Stack & Workflow Tools

Mobile Application (/mobile)

  • Framework: React Native SDK 54 (expo@~54.0.0, react-native@0.81.5, react@19.1.0)
  • Navigation: @react-navigation/native & @react-navigation/bottom-tabs
  • Animations: react-native-reanimated (UI-thread liquid flow spring animations)
  • Audio & Voice: expo-speech (Live Voice Coach TTS & Cadence Soundscapes)
  • Sharing & Files: expo-sharing & expo-file-system (GPX route export)
  • Data Visualization: react-native-gifted-charts & react-native-svg
  • Maps & GPS: react-native-maps & expo-location
  • Notifications: expo-notifications
  • Sensors & Pedometer: expo-sensors (Accelerometer & Pedometer with Rhythmic Cadence Filtering)

AI Backend (/backend)

Available in both Node.js / Express and Python / FastAPI:

  • Node.js: Express.js (ES Modules, CORS enabled)
  • Python: FastAPI, Pydantic, Uvicorn

πŸ“ Repository & File Structure

stepCounter/
β”œβ”€β”€ backend/                  # AI Analytics Engine Backend
β”‚   β”œβ”€β”€ app/
β”‚   β”‚   β”œβ”€β”€ analytics.js      # JS AI Analytics Engine logic
β”‚   β”‚   β”œβ”€β”€ analytics.py      # Python AI Analytics Engine logic
β”‚   β”‚   β”œβ”€β”€ main.js           # Express.js entry point & endpoints
β”‚   β”‚   β”œβ”€β”€ main.py           # FastAPI entry point & routes
β”‚   β”‚   └── models.py         # Pydantic schema models
β”‚   β”œβ”€β”€ package.json
β”‚   └── requirements.txt
β”‚
└── mobile/                   # React Native Mobile App (Expo SDK 54)
    β”œβ”€β”€ mocks/                # HealthKit / Pedometer mock module
    β”œβ”€β”€ src/
    β”‚   β”œβ”€β”€ components/       # UI Components
    β”‚   β”‚   β”œβ”€β”€ BarChart.tsx          # 30-day interactive bar chart
    β”‚   β”‚   β”œβ”€β”€ HealthSyncBanner.tsx  # HealthKit / Health Connect sync banner
    β”‚   β”‚   β”œβ”€β”€ InsightCard.tsx       # AI insight card renderer
    β”‚   β”‚   β”œβ”€β”€ LiquidScreenWrapper.tsx# Fluid liquid tab focus screen wrapper
    β”‚   β”‚   β”œβ”€β”€ LiquidSection.tsx     # Staggered wave section container
    β”‚   β”‚   β”œβ”€β”€ PermissionGate.tsx    # Permission gate wrapper
    β”‚   β”‚   β”œβ”€β”€ SplitTableCard.tsx    # Weekly pace comparison split card
    β”‚   β”‚   β”œβ”€β”€ StepRing.tsx          # Holographic concentric hero dial
    β”‚   β”‚   └── WidgetPreviewCard.tsx # System Widgets & Live Activity controller
    β”‚   β”œβ”€β”€ hooks/            # Custom React Hooks
    β”‚   β”‚   β”œβ”€β”€ useAnalytics.ts       # AI analytics hook
    β”‚   β”‚   β”œβ”€β”€ useStepHistory.ts     # Step history state manager
    β”‚   β”‚   └── useSteps.ts           # Sensor & hybrid pedometer hook
    β”‚   β”œβ”€β”€ screens/          # Main Tab & Feature Screens
    β”‚   β”‚   β”œβ”€β”€ BadgesScreen.tsx      # Gamified achievements & milestone badges
    β”‚   β”‚   β”œβ”€β”€ HistoryScreen.tsx     # 7-day selector & 30-day analytics history
    β”‚   β”‚   β”œβ”€β”€ HomeScreen.tsx        # Main dashboard screen
    β”‚   β”‚   β”œβ”€β”€ PermissionScreen.tsx   # Sensor & location permissions screen
    β”‚   β”‚   β”œβ”€β”€ PlannerScreen.tsx     # AI health planner & route bookmarking
    β”‚   β”‚   └── TrackerScreen.tsx     # Live GPS route map, audio coach, & weather
    β”‚   β”œβ”€β”€ services/         # App Services & Integrations
    β”‚   β”‚   β”œβ”€β”€ apiService.ts         # Backend API client & offline fallback engine
    β”‚   β”‚   β”œβ”€β”€ badgeService.ts       # Achievement & badge evaluation service
    β”‚   β”‚   β”œβ”€β”€ favoriteTrailsService.ts# Bookmarked favorite routes manager
    β”‚   β”‚   β”œβ”€β”€ gpxService.ts         # GPX route XML export & sharing
    β”‚   β”‚   β”œβ”€β”€ gpsService.ts         # Location tracking & route management
    β”‚   β”‚   β”œβ”€β”€ healthService.ts      # Native health platform sync bridge
    β”‚   β”‚   β”œβ”€β”€ notificationService.ts# Local push notification scheduling
    β”‚   β”‚   β”œβ”€β”€ soundscapeService.ts  # Cadence soundscapes & dynamic tempo engine
    β”‚   β”‚   β”œβ”€β”€ voiceCoachService.ts  # Audio Voice Coach TTS announcements
    β”‚   β”‚   β”œβ”€β”€ weatherService.ts     # Outdoor weather & AQI telemetry
    β”‚   β”‚   └── widgetService.ts      # Widget configuration service
    β”‚   β”œβ”€β”€ theme/            # Styling & Color System
    β”‚   β”‚   └── colors.ts             # OLED Cyber Dark color palette
    β”‚   └── utils/            # Helper Utilities
    β”‚       └── normalize.ts          # Responsive font & layout normalization
    β”œβ”€β”€ App.tsx               # Root App Component & Navigation Stack
    β”œβ”€β”€ app.json              # Expo application configuration
    β”œβ”€β”€ babel.config.js       # Babel compiler configuration
    β”œβ”€β”€ index.js              # Entry point registration
    β”œβ”€β”€ metro.config.js       # Metro bundler config
    β”œβ”€β”€ package.json          # Node dependencies & scripts
    └── tsconfig.json         # TypeScript configuration

βš™οΈ Quick Start Guide

1. Prerequisites

  • Node.js: >= 20.x
  • Expo Go app installed on your physical mobile phone (iOS / Android) or simulator/emulator.

πŸš€ Step-by-Step Execution Workflow

Step 1: Clone & Setup

git clone https://github.com/Ayush08k/VibeWalk.git
cd VibeWalk

Step 2: Start the Backend Server (Optional)

Option A: Node.js / Express Backend

cd backend
npm install
npm run dev

Server runs at http://localhost:8000

Option B: Python / FastAPI Backend

cd backend
python -m venv venv

# On Windows:
.\venv\Scripts\activate
# On macOS / Linux:
source venv/bin/activate

pip install -r requirements.txt
uvicorn app.main:app --reload --port 8000

API interactive documentation available at http://localhost:8000/docs


Step 3: Launch the React Native Mobile Application

cd mobile
npm install
npx expo start -c

Running Options:

  1. Physical Device (Recommended):
    • Open Expo Go on your iOS or Android phone.
    • Scan the QR code printed in the terminal.
  2. Android Emulator: Press a in the terminal after launching Expo CLI.
  3. iOS Simulator (macOS only): Press i in the terminal.

πŸ“œ License

Distributed under the MIT License.

About

VibeWalk a digital step counter application with AI integration for Multiple Sections

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages