Picovoice Platform
Flutter Quick Start

Platforms

• Flutter (1.20.0+)
• Android (4.4+, API 19+)
• iOS (9.0+)

Requirements

• Flutter SDK
• Android SDK (16+)
• JDK (8+)
• Xcode (9+)

Picovoice Account & AccessKey

Signup or Login to Picovoice Console to get your AccessKey. Make sure to keep your AccessKey secret.

Quick Start

Setup

1. Install Flutter SDK.

2. Run flutter doctor to determine any missing requirements.

3. Add the Picovoice plugin to your app project by referencing it in pubspec.yaml:

dependencies:
  picovoice_flutter: ^<version>

4. Enable the proper permission for recording with the hardware's microphone on both iOS and Android:

<key>NSMicrophoneUsageDescription</key>
<string>[Permission explanation]</string>

<uses-permission android:name="android.permission.RECORD_AUDIO"/>
<uses-permission android:name="android.permission.INTERNET"/>

Usage

1. Add a Porcupine keyword file (.ppn) and a Rhino context file (.rhn) to the assets folder in the project directory.

2. Add it to the pubspec.yaml:

flutter:
  assets:
    - assets/${KEYWORD_FILE}
    - assets/${CONTEXT_FILE}

3. Create an instance of PicovoiceManager using the constructor PicovoiceManager.create() that detects the wake word and infers intents from spoken commands:

import 'package:picovoice/picovoice_manager.dart';
import 'package:picovoice/picovoice_error.dart';

_picovoiceManager = PicovoiceManager.create(
    "{ACCESS_KEY}",
    "assets/${KEKWORD_FILE}",
    _wakeWordCallback,
    "assets/${CONTEXT_FILE}",
    _inferenceCallback);

Alternatively, if the model files are deployed to the device with a different method, the absolute paths to the files on device can be used.

The _wakeWordCallback and _inferenceCallback parameters are functions that are invoked when Porcupine detects the wake word and Rhino makes an intent inference, respectively.

void _wakeWordCallback(){    
    // wake word detected    
}

void _infererenceCallback(RhinoInference inference){
    if(inference.isUnderstood!){
        String intent = inference.intent!
        Map<String, String> slots = inference.slots!
        // take action based on inferred intent and slot values
    }
    else{
        // handle unsupported commands
    }    
}

Start audio capture and processing:

try{
    await _picovoiceManager.start();
} on PicovoiceException catch(ex){
    // deal with Picovoice init error
}

Stop it when done with Picovoice:

await _picovoiceManager.stop();

Custom Wake Words & Contexts

1. Create custom wake words and contexts with the Picovoice Console.

2. Download the custom Porcupine keyword (.ppn) and Rhino context (.rhn) files.

3. Add them to the pubspec.yaml:

flutter:
  assets:
    - assets/${CUSTOM_KEYWORD_FILE}
    - assets/${CUSTOM_CONTEXT_FILE}

4. Create an instance of PicovoiceManager using the constructor PicovoiceManager.create().

Non-English Languages

Use the corresponding model file (.pv) to infer non-English wake words and contexts. The model files for all supported languages are available on the Porcupine GitHub repository and Rhino GitHub repository.

1. Add the model files to the assets folder in the project directory as well as keyword and context files.
2. Add it to the pubspec.yaml:

flutter:
  assets:
    - assets/${CUSTOM_KEYWORD_FILE}
    - assets/${CUSTOM_CONTEXT_FILE}
    - assets/${PORCUPINE_MODEL_FILE}
    - assets/${RHINO_MODEL_FILE}

3. Create an instance of PicovoiceManager using the constructor PicovoiceManager.create():

_picovoiceManager = PicovoiceManager.create(
    "{ACCESS_KEY}",
    "assets/{CUSTOM_KEKWORD_FILE}",
    _wakeWordCallback,
    "assets/{CUSTOM_CONTEXT_FILE}",
    _inferenceCallback,
    porcupineModelPath="assets/{PORCUPINE_MODEL_FILE}",
    rhinoModelPath="assets/{RHINO_MODEL_FILE}");

Alternatively, if the model files are deployed to the device with a different method, the absolute paths to the files on device can be used.

Demo

For the Picovoice Flutter SDK, we offer demo applications that demonstrate how to use the End-to-End Picovoice platform on real-time audio streams (i.e. microphone input).

Setup

Clone the Picovoice GitHub Repository:

git clone --recurse-submodules https://github.com/Picovoice/picovoice.git

Usage

1. Replace {YOUR_ACCESS_KEY_HERE} with a valid AccessKey in the demo/flutter/lib/main.dart file:

2. Copy assets:

cd picovoice/demo/flutter
bash copy_assets.sh

NOTE: on Windows, Git Bash or another bash shell is required, or you will have to manually copy

• The android keyword into ${DEMO_FOLDER}/assets/keyword_files/android.
• The android context into ${DEMO_FOLDER}/assets/contexts/android.
• The iOS keyword into ${DEMO_FOLDER}/assets/keyword_files/ios.
• The iOS context into ${DEMO_FOLDER}/assets/contexts/ios.

3. Build and deploy the demo to your device:

flutter run

For more information on our Picovoice demos for Flutter, head over to our GitHub repository.

Resources