ahmedabdelkarim/VoiceControlKit

iOS framework that enables detecting and handling voice commands using microphone.

VoiceControlKit

iOS framework that enables detecting and handling voice commands using microphone. Built using Swift with minumum target iOS 14.3.

Features

• Detects commands of single word ("Open", "Close"), and sentences ("Show first item details", "Go back").
• Can work both online and offline, with extremely fast response time (< 0.5 second).
• Ability to use different set of commands for each screen.
• Works without Siri integration. So, users don't have to say "Hey Siri" to detect commands (100% free style detection).
• Very simple to configure and use.
• Low Memory Footprint (~0.3MBs)
• Low CPU consumption (~2% without speech, ~3% while speeking) [% of single core not all processor cores]

Prerequisites

• macOS Big Sur, or later
• Xcode 12.3+
• iPhone device with iOS 14.3+

Install VoiceControlKit (using CocoaPods)

1. Make sure you have CocoaPods installed.
2. Update local pod repo using command pod repo update or pod repo update trunk.
3. Open Terminal from your project folder, and run commad pod init.
4. Add pod 'VoiceControlKit' inside Podfile, and run pod install.

Configure iOS Project

1. Open Info.plist and add description for keys NSSpeechRecognitionUsageDescription and NSMicrophoneUsageDescription.
2. Use VoiceCommandListener and VoiceCommandListenerDelegate in your code (see below example) to detect and handle voice commands.
3. Test on a real iPhone device (not simulator).
4. Make sure Siri is enabled on iPhone device from Settings, Siri & Search, Listen for "Hey Siri".

Code Example

import UIKit
import VoiceControlKit

class ViewController: UIViewController, VoiceCommandListenerDelegate {
    // MARK: - Outlets
    @IBOutlet weak var label: UILabel!
    
    // MARK: - Lifecycle
    override func viewDidLoad() {
        super.viewDidLoad()
        
        let commands = [
            VoiceCommand(text: "Hello"),
            VoiceCommand(text: "Open next page"),
            VoiceCommand(text: "Take photo"),
            VoiceCommand(text: "Go back"),
            VoiceCommand(text: "Close app")
        ]
        
        VoiceCommandListener.shared.delegate = self
        
        VoiceCommandListener.shared.start(with: commands, success: {
            // Now it's listening to voice commands
        }, failure: { error in
            print(error.localizedDescription)
        })
    }
    
    // MARK: - VoiceCommandListenerDelegate
    func voiceCommandListener(_ listener: VoiceCommandListener, detected command: VoiceCommand) {
        
        // Do whatever you want depending on detected command
    
        DispatchQueue.main.async {
            self.label.text = command.text
        }
    }
}