MoodFlow

A web app that tracks your mood during focus sessions. Record a short voice check-in at the start and end of each session. The app analyzes your emotional state using AI and shows you patterns over time.

Features

Voice Check-ins — Record yourself talking at the beginning and end of a focus session
Emotion Analysis — Audio analyzed using VAD (Valence-Arousal-Dominance) deep learning model
3D Dashboard — Interactive visualization with live emotion timeline
Insights & Recommendations — Focus stability metrics, emotion distribution, and AI-generated suggestions
Session Management — Browse, filter, and search through all past sessions

Architecture

moodflow/
├── web/                    # Next.js 14 frontend
│   ├── app/               # Pages (dashboard, sessions, insights, login)
│   ├── components/        # React components with Framer Motion
│   ├── lib/               # Utilities (API client, analytics, recording)
│   └── prisma/            # Database schema
├── ai-service/            # FastAPI AI inference service
│   ├── app/
│   │   ├── pipelines/     # ML inference pipelines
│   │   ├── routes/       # FastAPI endpoints
│   │   ├── cache/        # Model caching system
│   │   └── utils/        # Audio preprocessing
│   └── scripts/           # Management scripts
└── docker-compose.yml     # PostgreSQL + MinIO

Tech Stack

Layer	Technologies
Frontend	Next.js 14, React 18, TypeScript, Tailwind CSS, Framer Motion
3D / Visual	Three.js, React Three Fiber
Backend	Next.js API Routes
Database	PostgreSQL + Prisma ORM
File Storage	MinIO (S3-compatible)
AI Service	FastAPI, PyTorch, Transformers, Whisper, Wav2Vec2

Overview

MoodFlow is a modern web application that helps users track their emotional well-being during work sessions. Through voice check-ins and AI-powered analysis, it provides deep insights into mood patterns, focus stability, and stress levels.

Features

Voice Check-ins: Record voice notes at the start and end of focus sessions
AI-Powered Analysis: Emotion detection using VAD (Valence-Arousal-Dominance) model
Real-time Dashboard: Live tracking of sessions, focus time, and current mood
Insights & Trends: Weekly emotional patterns, focus stability metrics, and AI recommendations
Session Management: Create, track, and analyze focus sessions with detailed analytics

Tech Stack

Component	Technology
Frontend	Next.js 14, React 18, TypeScript
Styling	Tailwind CSS, Framer Motion
3D Graphics	Three.js, React Three Fiber
Backend	Next.js API Routes
Database	PostgreSQL with Prisma ORM
Storage	MinIO (S3-compatible)
AI Service	FastAPI, PyTorch, Transformers, Whisper

Quick Start

1. Start Infrastructure

docker-compose up -d

PostgreSQL: localhost:5432
MinIO API: localhost:9000
MinIO Console: localhost:9001

2. Web App Setup

This starts PostgreSQL (port 5432) and MinIO (ports 9000, 9001).

2. Start AI Service

cd ai-service
pip install -r requirements.txt
uvicorn app.main:app --host 0.0.0.0 --port 8000

3. Start Web Application

cd web
npm install
cp .env.example .env.local
npm run prisma:generate
npm run prisma:migrate
npm run dev

Open http://localhost:3000

3. AI Service Setup

cd ai-service
pip install -r requirements.txt

# Pre-warm model cache (downloads ML models)
python scripts/cache_models.py

# Start the service
uvicorn app.main:app --host 0.0.0.0 --port 8000

Open http://localhost:8000/docs for API documentation.

AI Service

The AI service provides emotion detection and speech-to-text transcription through REST APIs.

Endpoints

Endpoint	Method	Description
`/health`	GET	Health check
`/process-checkin/`	POST	Short audio check-in processing
`/analyze-audio/`	POST	Long audio analysis with stress tracking

AI Pipeline

Audio Input (16kHz mono)
       ↓
┌─────────────────────────┐
│  Wav2Vec2 Feature      │
│  Extractor             │
│  (audeering model)     │
└─────────────────────────┘
       ↓
┌─────────────────────────┐
│  Pre-trained Wav2Vec2  │
│  Emotion Detection      │
│  (VAD Regression)       │
└─────────────────────────┘
       ↓
┌─────────────────────────┐
│  Stress Calculation    │
│  + Emotion Mapping      │
└─────────────────────────┘
       ↓
Output: { valence, arousal, dominance, stress_index, calm, stress, focus }

Models Used

Model	Purpose	Size
`audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim`	Emotion detection (VAD)	~800MB
`openai/whisper-base`	Speech-to-text transcription	~300MB

Model Caching

Models are cached locally to avoid re-downloading on every cold start:

# Pre-warm cache (run during deployment)
python scripts/cache_models.py

# Check cache status
python scripts/cache_models.py --info

# Verify cache integrity
python scripts/cache_models.py --verify

Environment Variables

# AI Service (.env)
GEMINI_API_KEY=your_api_key          # For topic extraction
MODEL_CACHE_DIR=./models              # Local model cache directory
USE_MODEL_CACHE=true                  # Enable/disable caching

Web App

Available Scripts

cd web

npm run dev           # Development server
npm run build         # Production build
npm run start         # Production server
npm run lint          # ESLint
npm run prisma:generate   # Generate Prisma client
npm run prisma:migrate   # Run migrations
npm run prisma:push      # Push schema (no migration history)

Environment Variables

Variable	Default	Description
`DATABASE_URL`	`postgresql://...`	PostgreSQL connection string
`AI_SERVICE_BASE_URL`	`http://localhost:8000`	AI service URL
`MINIO_ENDPOINT`	`localhost`	MinIO server
`MINIO_PORT`	`9000`	MinIO port
`MINIO_ACCESS_KEY`	`minio_user`	MinIO access key
`MINIO_SECRET_KEY`	`minio_password`	MinIO secret key
`MINIO_BUCKET`	`moodflow`	MinIO bucket name

Database

Schema Overview

User — Account info and preferences
FocusSession — Focus session with start/end emotions and metrics
Recording — Audio files with analysis results
UserPreference — Per-user settings (theme, reminders, audio quality)

Development

Running Tests

# Web app
cd web && npm test

# AI service
cd ai-service && pytest

Building Docker Images

# Build web app
cd web && docker build -t moodflow-web .

# Build AI service
cd ai-service && docker build -t moodflow-ai .

License

MIT

Environment Variables

Web Application (`web/.env.local`)

Variable	Description	Default
`DATABASE_URL`	PostgreSQL connection string	`postgresql://moodflow_user:moodflow_password@localhost:5432/moodflow_db`
`MINIO_ENDPOINT`	MinIO server endpoint	`localhost`
`MINIO_PORT`	MinIO server port	`9000`
`MINIO_ACCESS_KEY`	MinIO access key	`minio_user`
`MINIO_SECRET_KEY`	MinIO secret key	`minio_password`
`MINIO_BUCKET`	MinIO bucket name	`moodflow`
`AI_SERVICE_URL`	AI service URL	`http://localhost:8000`

AI Service (`ai-service/.env`)

Create based on .env.example in the ai-service directory.

Project Structure

moodflow/
├── docker-compose.yml     # PostgreSQL + MinIO
├── web/                   # Next.js application
│   ├── app/               # App router pages
│   │   ├── api/           # API routes
│   │   ├── dashboard/     # Dashboard pages
│   │   └── login/         # Authentication
│   ├── components/        # React components
│   ├── lib/               # Utility functions
│   └── prisma/            # Database schema
├── ai-service/            # FastAPI AI service
│   ├── app/
│   │   ├── pipelines/    # Audio processing pipelines
│   │   ├── routes/       # API endpoints
│   │   └── utils/        # Utility functions
│   └── requirements.txt
└── LICENSE

Running Services

Service	Command	URL
PostgreSQL	`docker-compose up -d postgres`	localhost:5432
MinIO	`docker-compose up -d minio`	localhost:9000 (API), localhost:9001 (Console)
AI Service	`uvicorn app.main:app --host 0.0.0.0 --port 8000`	localhost:8000
Web App	`npm run dev` (in web/)	localhost:3000

Available Scripts

Web Application

Script	Description
`npm run dev`	Start development server
`npm run build`	Build for production
`npm run start`	Start production server
`npm run lint`	Run ESLint
`npm run prisma:generate`	Generate Prisma client
`npm run prisma:migrate`	Run database migrations
`npm run prisma:push`	Push schema to database

License

This project is licensed under the MIT License - see the LICENSE file for details.

Author

paledreamer - GitHub
Praveen Parakh - GitHub

7fe3eba7b873f1d3c5064aece88ccc06fe2708ea

Name		Name	Last commit message	Last commit date
ai-service		ai-service
web		web
.dockerignore		.dockerignore
.gitignore		.gitignore
Dockerfile.android		Dockerfile.android
LICENSE		LICENSE
README.md		README.md
docker-compose.yml		docker-compose.yml
package-lock.json		package-lock.json
package.json		package.json

Latest commit

History

Folders and files

Repository files navigation

MoodFlow

Features

Architecture

Tech Stack

Overview

Features

Tech Stack

Quick Start

1. Start Infrastructure

2. Web App Setup

2. Start AI Service

3. Start Web Application

3. AI Service Setup

AI Service

Endpoints

AI Pipeline

Models Used

Model Caching

Environment Variables

Web App

Available Scripts

Environment Variables

Database

Schema Overview

Development

Running Tests

Building Docker Images

License

Environment Variables

Web Application (web/.env.local)

AI Service (ai-service/.env)

Project Structure

Running Services

Available Scripts

Web Application

License

Author

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages

Web Application (`web/.env.local`)

AI Service (`ai-service/.env`)