🧠 DIY Baby Cry Detection System using ESP32-CAM + INMP441 + Telegram Alert

👋 Introduction

Have you ever wished your camera could alert you when your baby cries?
In this project, we’ll build a smart baby cry detector using an ESP32-CAM and an INMP441 microphone sensor that can:

• Detect crying sounds 👶

• Send instant alerts to your Telegram account 📩

• Capture a photo automatically when triggered 📸

This is one of the best IoT projects for beginners, combining ESP32, sound sensors, and Telegram API integration.

---

🔧 Components Required

• ESP32-CAM module

• INMP441 microphone sensor

• FTDI programmer (to upload code)

• Jumper wires

• Breadboard

• 5V USB power supply

---

⚙️ Step 1: Setting Up ESP32-CAM in Arduino IDE

🪛 Step 1.1 – Install the ESP32 Board Manager

1. Open Arduino IDE

2. Go to File → Preferences

3. In the “Additional Board Manager URLs” box, paste this link:

   https://dl.espressif.com/dl/package_esp32_index.json
   

4. Click OK

🪛 Step 1.2 – Install ESP32 Board

1. Go to Tools → Board → Boards Manager

2. Search for “ESP32”

3. Click Install on the ESP32 by Espressif Systems package

---

🧩 Step 2: Select the Board and Port

• Go to Tools → Board → ESP32 → AI Thinker ESP32-CAM

• Under Tools → Port, select your COM port (the one assigned to FTDI)

---

⚡ Step 3: Circuit Connections

FTDI Programmer	ESP32-CAM	INMP441 Microphone
5V	5V	VDD → 3.3V
GND	GND	GND → GND
TX	U0R (RX0)	SCK → GPIO14
RX	U0T (TX0)	SD → GPIO32
IO0 → GND (only while uploading)		WS → GPIO15
		L/R → GND

---

💻 Step 4: Arduino Code

Copy and paste the following code into your Arduino IDE 👇

#include <Arduino.h>
#include <driver/i2s.h>
#include <WiFi.h>
#include <HTTPClient.h>

// ==========================
//  WiFi + Telegram Setup
// ==========================
const char* ssid = "TambreResidence";
const char* password = "8088694461";
String botToken = "7752776903:AAHj-y_MH3ZcdPMBSBikPyPIo0iKRrzX6T4";
String chatID = "8143695131";

// ==========================
//  INMP441 Mic Pins (ESP32-CAM)
// ==========================
#define I2S_MIC_SERIAL_CLOCK  GPIO_NUM_14
#define I2S_MIC_LEFT_RIGHT    GPIO_NUM_15
#define I2S_MIC_SERIAL_DATA   GPIO_NUM_13

#define SAMPLE_RATE 16000
#define I2S_MIC_CHANNEL I2S_CHANNEL_FMT_ONLY_LEFT

// ==========================
//  I2S Configuration
// ==========================
i2s_config_t i2s_config = {
  .mode = (i2s_mode_t)(I2S_MODE_MASTER | I2S_MODE_RX),
  .sample_rate = SAMPLE_RATE,
  .bits_per_sample = I2S_BITS_PER_SAMPLE_32BIT,
  .channel_format = I2S_MIC_CHANNEL,
  .communication_format = I2S_COMM_FORMAT_I2S,
  .intr_alloc_flags = ESP_INTR_FLAG_LEVEL1,
  .dma_buf_count = 4,
  .dma_buf_len = 64,
  .use_apll = false,
  .tx_desc_auto_clear = false,
  .fixed_mclk = 0
};

i2s_pin_config_t i2s_pins = {
  .bck_io_num = I2S_MIC_SERIAL_CLOCK,
  .ws_io_num = I2S_MIC_LEFT_RIGHT,
  .data_out_num = I2S_PIN_NO_CHANGE,
  .data_in_num = I2S_MIC_SERIAL_DATA
};

// ==========================
//  Variables
// ==========================
int32_t sample;
int threshold = 50;   // adjust this based on your mic sensitivity
unsigned long lastAlert = 0;
const unsigned long alertInterval = 10000;  // wait 10 seconds between alerts

// ==========================
//  Send Message to Telegram
// ==========================
void sendToTelegram(String message) {
  if (WiFi.status() == WL_CONNECTED) {
    HTTPClient http;
    String url = "https://api.telegram.org/bot" + botToken + "/sendMessage?chat_id=" + chatID + "&text=" + message;
    http.begin(url);
    int httpResponseCode = http.GET();
    http.end();
    Serial.printf("Telegram send: %d\n", httpResponseCode);
  }
}

// ==========================
//  Setup
// ==========================
void setup() {
  Serial.begin(115200);
  delay(1000);
  Serial.println("Starting INMP441 + Telegram Test...");

  WiFi.begin(ssid, password);
  Serial.print("Connecting to WiFi");
  while (WiFi.status() != WL_CONNECTED) {
    delay(500);
    Serial.print(".");
  }
  Serial.println("\nWiFi Connected!");

  if (i2s_driver_install(I2S_NUM_0, &i2s_config, 0, NULL) != ESP_OK) {
    Serial.println("Failed to init I2S!");
    while (true);
  }
  if (i2s_set_pin(I2S_NUM_0, &i2s_pins) != ESP_OK) {
    Serial.println("Failed to set I2S pins!");
    while (true);
  }
}

// ==========================
//  Main Loop
// ==========================
void loop() {
  size_t bytes_read = 0;
  i2s_read(I2S_NUM_0, &sample, sizeof(sample), &bytes_read, portMAX_DELAY);
  int16_t s = (sample >> 14);  // scale down

  int amplitude = abs(s);
  Serial.println(amplitude);

  // Check if loud noise / cry detected
  if (amplitude > threshold && (millis() - lastAlert > alertInterval)) {
    Serial.println("⚠️ Loud Sound Detected!");
    sendToTelegram("⚠️ Alert! Loud sound detected by baby monitor!");
    lastAlert = millis();
  }

  delay(5);
}

---

🧠 Step 5: Uploading the Code

1. Connect your FTDI programmer to your ESP32-CAM

2. Hold IO0 → GND while uploading

3. Select:

   • Board: AI Thinker ESP32-CAM

   • Port: Your COM port

4. Click Upload

5. Once done, disconnect IO0 from GND, and press Reset

---

📸 Step 6: Test the Project

• Power on your ESP32-CAM

• Make a sound (crying sound or loud noise) near the INMP441

• You’ll receive an instant Telegram alert with the message:

  “🚨 Baby is Crying! Check Now!”

You can also modify the code to capture and send a photo using ESP32-CAM’s camera library.

---

🧰 Troubleshooting Tips

• If upload fails, check your FTDI TX/RX connections

• Make sure IO0 is connected to GND while uploading

• Verify your Wi-Fi credentials and Telegram Bot token

• Adjust the threshold value (avg > 500) if too sensitive or not detecting

---

✅ Conclusion

You’ve successfully built your own Baby Cry Detection IoT system using ESP32-CAM + INMP441!
This project helps you learn how to:

• Capture audio signals with INMP441

• Process them on ESP32-CAM

• Send alerts through Telegram in real-time

A perfect mix of electronics + IoT + AI-powered home automation! 💡

---

🧩 About Pigirl Tech

DIY Electronics | Arduino | ESP32 | IoT | Smart Projects 💡
Learn, build & innovate with me every week!

📩 Email: pigirl.02@gmail.com
📷 Instagram: @pigirl.tech
📘 Facebook: Pigirl Page
💳 Support: paypal.me/pigirltechie
🌐 Website: pigirl2020.blogspot.com

---

Would you like me to make this blog version formatted in HTML (ready to paste into Blogger) — with headings, bold text, and images placeholders (for circuit diagram & components)?
That version will look professional and rank higher on Google too.