Language detection guide for Android

The MediaPipe Language Detector task lets you identify the language of a piece of text. These instructions show you how to use the Language Detector with Android apps. The code sample described in these instructions is available on GitHub.

You can see this task in action by viewing the demo. For more information about the capabilities, models, and configuration options of this task, see the Overview.

Code example

The example code for Language Detector provides a simple implementation of this task for your reference. This code help you test this task and get started on building your own language detection feature. You can browse the Language Detector example code on GitHub.

Download the code

The following instructions show you how to create a local copy of the example code using the git version control command line tool.

To download the example code:

1. Clone the git repository using the following command:

   git clone https://github.com/google-ai-edge/mediapipe-samples
   

2. Optionally, configure your git instance to use sparse checkout, so you have only the files for the Language Detector example app:

   cd mediapipe
   git sparse-checkout init --cone
   git sparse-checkout set examples/languagedetector/android
   

For instruction on how to setup and run an example with Android Studio, see the example code setup instructions in the Setup Guide for Android.

Key components

The following files contain the crucial code for the text classification example app:

• LanguageDetectorHelper.kt - Initializes the language detector and handles the model selection.
• ResultsAdapter.kt - Handles and formats the detection results.
• MainActivity.kt - Implements the application, including calling LanguageDetectorHelper and ResultsAdapter.

Setup

This section describes key steps for setting up your development environment and code projects specifically to use Language Detector. For general information on setting up your development environment for using MediaPipe tasks, including platform version requirements, see the Setup guide for Android.

Dependencies

Language Detector uses the com.google.mediapipe:tasks-text libraries. Add this dependency to the build.gradle file of your Android app development project. You can import the required dependencies with the following code:

dependencies {
    implementation 'com.google.mediapipe:tasks-text:latest.release'
}

Model

The MediaPipe Language Detector task requires a trained model that is compatible with this task. For more information on available trained models for Language Detector, see the task overview Models section.

Select and download the model, and then store it within your project directory:

<dev-project-root>/src/main/assets

Specify the path of the model within the ModelName parameter.

Create the task

You can use one of the createFrom...() functions to create the task. The createFromOptions() function accepts configuration options for the language detector. You can also initialize the task using the createFromFile() factory function. The createFromFile() function accepts a relative or absolute path to the trained model file. For more information on configuring tasks, see Configuration options.

The following code demonstrates how to create and configure this task.

// For creating a language detector instance:
LanguageDetectorOptions options =
       LanguageDetectorOptions.builder()
       .setBaseOptions(
          BaseOptions.builder()
            .setModelAssetPath(modelPath)
            .build()
          )
       .build();
LanguageDetector languageDetector = LanguageDetector.createFromOptions(context, options);

You can see an example of how to create a task in the code example LanguageDetectorHelper class initDetector() function.

Configuration options

This task has the following configuration options for Android apps:

Prepare data

Language Detector works with text (String) data. The task handles the data input preprocessing, including tokenization and tensor preprocessing. All preprocessing is handled within the detect() function. There is no need for additional preprocessing of the input text beforehand.

String inputText = "Some input text for the language detector";

Run the task

The Language Detector uses the LanguageDetector.detect() method to process input text and predict the language of the text. You should use a separate execution thread for executing the detection to avoid blocking the Android user interface thread with your app.

The following code demonstrates how to execute the processing with the task model using a separate execution thread.

// Predict the language of the input text.
fun classify(text: String) {
    executor = ScheduledThreadPoolExecutor(1)

    executor.execute {
        val results = languageDetector.detect(text)
        listener.onResult(results)
    }
}

You can see an example of how to run a task in the code example LanguageDetectorHelper class detect() function.

Handle and display results

The Language Detector outputs a LanguageDetectorResult consisting of a list of language predictions along with the probabilities for those predictions. The language categories are defined in the model, see the task overview Models section for details on the model you are using.

The following shows an example of the output data from this task:

LanguageDetectorResult:
  LanguagePrediction #0:
    language_code: "fr"
    probability: 0.999781

This result has been obtained by running the model on the input text: "Il y a beaucoup de bouches qui parlent et fort peu de têtes qui pensent.".

You can see an example of how to display results in the code example ResultsAdapter class and ViewHolder inner class.