@empathai/core

EmpathAI Core — lightweight, non-intrusive emotion detection for web apps (keyboard + mouse).

What it does

empathai-core analyzes user interaction signals (mouse movement, typing rhythm, clicks) and returns inferred emotional states (e.g. focused, frustrated, bored, neutral) together with a basic confidence score.
It is intentionally privacy-first — no camera/mic required.

What EmpathAI Core Does

• Captures non-intrusive user behavior signals
• Infers basic emotional states (deterministic v1)
• Emits structured emotion data with confidence & timestamp
• Works entirely on the client
• Stores no personal data
• Safe for BFSI / SaaS / FinTech / enterprise / privacy-sensitive environments

What EmpathAI Core Does NOT Do

• No camera access
• No microphone access
• No personal data collection

EmpathAI Core provides signals, not decisions.

Installation

npm install empathai-core
# or
pnpm add empathai-core

Usage Examples

JavaScript (Framework-Agnostic)

import { createEmpathAI } from 'empathai-core';

const empathAI = createEmpathAI({
  onEmotionDetected: (emotion) => {
    console.log('Emotion detected:', emotion);
  },
  // Optional configuration
  signalWindowMs: 3000,      // 3 second window
  analysisIntervalMs: 1000,  // Analyze every second
  mouseThrottleMs: 50,       // Throttle mouse events
  confidenceThreshold: 0.3,  // Minimum confidence
  debugMode: false,          // Enable for debugging
});

empathAI.init();

// Optional cleanup
window.addEventListener('beforeunload', () => {
  empathAI.destroy();
});

Java Backend with Web Frontend

Frontend (JavaScript)

import { createEmpathAI } from 'empathai-core';

const empathAI = createEmpathAI({
  onEmotionDetected: (emotion) => {
    fetch('/api/emotion', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify(emotion),
    });
  }
});

empathAI.init();

Backend (Spring Boot example)

@RestController
@RequestMapping("/api")
public class EmotionController {

    @PostMapping("/emotion")
    public ResponseEntity<Void> receiveEmotion(@RequestBody EmotionPayload payload) {
        // Store, analyze, or react to emotion signal
        return ResponseEntity.ok().build();
    }
}

Python Backend with Web Frontend

Frontend (JavaScript)

import { createEmpathAI } from 'empathai-core';

const empathAI = createEmpathAI({
  onEmotionDetected: (emotion) => {
    fetch('/emotion', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify(emotion),
    });
  }
});

empathAI.init();

Backend (Flask example)

from flask import Flask, request

app = Flask(__name__)

@app.route("/emotion", methods=["POST"])
def receive_emotion():
    data = request.json
    # Process emotion signal
    return "", 200

License

MIT