Face Recognition Smart Door with Voice Greeting — No Code

What is this project about?

Build a Face Recognition Smart Door that registers faces, identifies them in real time, and controls a servo motor to open and close a door—complete with personalized voice greetings powered by text-to-speech. Walk up to the camera, and if your face is registered, the system greets you by name and opens the door automatically.

Why did we decide to make it?

Building a face recognition door system traditionally requires deep learning frameworks, servo control libraries, audio synthesis, and a lot of glue code. With Grablo's visual blocks, all of that complexity disappears—no coding needed.

What You'll Build

A complete face recognition smart door featuring:

• Real-time face registration and recognition
• Servo motor door control (open/close)
• Personalized voice greetings via text-to-speech
• False-trigger prevention with hold-time logic
• Dashboard with live camera feed and enrollment controls

Demo

Quick Start

• Install software:Download and install Grablo on your Raspberry Pi
• Set up hardware: Connect camera, speaker, and servo motor
• Get this project:Project Link
• Open Grablo app: On your desktop or mobile browser
• Hit RUN: Connect to your Raspberry Pi and start recognizing faces!
• Enroll faces: Type a name and click the Enroll button
• Test it out: Walk up to the camera and hear your personalized greeting

💡 Want to build it yourself? Continue to Build Process

Build Process

Dashboard

Create Project & Dashboard

• Go to https://app.grablo.co
• Click "Create Project"
• Name your project
• Select your device (Raspberry Pi 4 for this tutorial)
• Create a new dashboard

Add Widgets

• Camera Widget: Select your camera to display the live video feed
• Push Button Widget: Create an "Enroll" variable to trigger face registration
• Input Widget: Create a "Name to Enroll" variable for typing the person's name
• Label Widget (Face Count): Display the total number of enrolled faces
• LED Widget (Registered): Indicate whether the detected face is a registered one
• Label Widget (Name): Display the recognized person's name
• LED Widget (Door Open): Show whether the door is currently open or closed

Logic

Control 1: Start Camera with Face Recognition

Start the camera and enable face recognition when the project begins.

• Condition (Once): Runs once at project start
• Action (AI Analysis): Create a new AI Analyzer → Set AI Model to Face Recognition → Type to Recognize Face → Assign output variables (Registered and Recognized Name)
• Action (Camera): Select camera → Command: Start

Control 2: Face Enrollment

Register a new face whenever the Enroll button is pressed on the dashboard.

• Condition (Compare): Enroll variable = True, Option = Rising Edge
• Action (AI Analysis): Select AI Analyzer → Command: Add Analysis → Type: Enroll Face → Face Name: From Variable → select "Name to Enroll"

Control 3: Enrolled Face Count

Continuously track the total number of registered faces.

• Condition (Always): Runs on every cycle
• Action (AI Analysis): Select AI Analyzer → Command: Add Analysis → Type: Get Enrolled Count → Assign to Face Count variable

Control 4: Door Opening

Open the door when a registered face is detected continuously for 3 seconds (prevents false triggers).

• Condition 1 (Compare): Registered = True, Option = Hold Time: 3 seconds
• Condition 2 (Compare): Door Open = False (only open if currently closed)
• Action (TTS): Create new TTS → Command: Play → Output Text: Use block editor to join "Welcome home, " + Recognized Name variable + ". Door opening."
• Action (IO Device): Create Servo Motor → Select GPIO pin → Command: Set Position → 90 degrees
• Action (Set Variable): Door Open = True

Control 5: Door Closing

Close the door when no registered face is detected for 3 seconds.

• Condition 1 (Compare): Registered = False, Option = Hold Time: 3 seconds
• Condition 2 (Compare): Door Open = True (only close if currently open)
• Action (TTS): Command: Play → Output Text: "Door closing."
• Action (IO Device): Select Servo Motor → Command: Set Position → 0 degrees
• Action (Set Variable): Door Open = False

💡 Tip: You can duplicate Control 4 and modify the conditions and actions to quickly create Control 5.

That's It!

Congratulations! You've built a complete face recognition smart door using visual blocks and zero lines of code.

Expected Results

When you run your project:

• Camera view appears: Live feed with face detection overlay
• Unregistered face: Red border labeled "Unknown"—door stays closed
• Enroll a face: Type a name, click Enroll → face count increases, face is recognized immediately
• Registered face detected for 3 seconds: TTS greets by name, servo rotates to 90°, door opens
• No registered face for 3 seconds: TTS says "Door closing", servo returns to 0°, door closes
• Scale: Register hundreds or even thousands of faces

Hardware Setup

Camera Options:

• Raspberry Pi Camera Module (CSI)
• USB Camera

Speaker Options:

• 3.5mm audio jack
• USB audio adapter
• Bluetooth speaker

Servo Motor Wiring:

Follow the connection table and connect each row of components horizontally:

⚠️ For multiple servos or heavy loads, use an external power supply instead of powering directly from the Raspberry Pi.

Video Tutorial

Watch the complete build process in action:

What's Next?

Expand your project:

• Add a buzzer alert for unregistered faces
• Log entry/exit times to a dashboard
• Add multiple doors with different servo motors
• Integrate with a relay to control an electric door lock
• Add a "Delete Face" button for managing enrolled faces

Troubleshooting

Face not detecting:

• Ensure adequate lighting on the face
• Check camera connection and orientation
• Make sure the face is within the camera's field of view

Door opens for unknown faces:

• Verify the Compare condition checks Registered = True
• Ensure Hold Time is set (3 seconds recommended)
• Check that the correct variable is assigned

TTS not speaking:

• Check speaker connection
• Verify TTS action is configured with Command: Play
• Test with a simple text first before using variables

Servo not moving:

• Verify GPIO pin matches your wiring
• Check power supply to the servo
• Test with different angle values (0° and 90°)

Door keeps opening and closing rapidly:

• Verify Hold Time is set in both open and close conditions
• Ensure Door Open variable is properly toggled in both controls
• Check that conditions include the door state check (open/closed)

Resources & Support

• Website: https://grablo.co
• Web App: https://app.grablo.co
• Docs: https://doc.grablo.co
• Support: support@grablo.co