Panlingo.LanguageIdentification.MediaPipe.Native 0.0.0.23

Panlingo.LanguageIdentification.MediaPipe

Welcome to Panlingo.LanguageIdentification.MediaPipe, a .NET wrapper for the MediaPipe library by Google Inc. This package seamlessly integrates language identification capabilities into .NET applications, enabling accurate and efficient recognition of over 107 languages with minimal effort. Perfect for applications dealing with multilingual texts or requiring automatic language detection.

Requirements

• Runtime: .NET >= 5.0
• OS: Linux
• Arch: AMD64

Installation

To integrate the MediaPipe functionality, follow these steps:

Install the NuGet package:

dotnet add package Panlingo.LanguageIdentification.MediaPipe

Usage

Integrating the MediaPipe library into your .NET application is straightforward. Here’s a quick guide to get you started:

1. Install the Package: Ensure you have added the Panlingo.LanguageIdentification.MediaPipe package to your project using the provided installation command.
2. Initialize the Library: Follow the example snippet to initialize and use the MediaPipe library for detecting languages.

using Panlingo.LanguageIdentification.MediaPipe;

class Program
{
    static void Main()
    {
        using var mediaPipe = new MediaPipeDetector(
            options: MediaPipeOptions.FromDefault()
        );

        var text = "Привіт, як справи?";

        var predictions = mediaPipe.PredictLanguages(text);

        foreach (var prediction in predictions)
        {
            Console.WriteLine(
                $"Language: {prediction.Language}, " +
                $"Probability: {prediction.Probability}"
            );
        }
    }
}

Custom models

Download the Pretrained Models

Download the pretrained language identification (LID) model provided by Google:

curl --location -o /models/mediapipe_language_detector.tflite https://storage.googleapis.com/mediapipe-models/language_detector/language_detector/float32/1/language_detector.tflite

Learn more about this model here:

• Google AI Edge

Use custom model in code

var modelPath = "/models/mediapipe_language_detector.tflite";
using var mediaPipe = new MediaPipeDetector(
    options: MediaPipeOptions.FromFile(modelPath)
);

OR

var modelPath = "/models/mediapipe_language_detector.tflite";
using var stream = File.Open(modelPath, FileMode.Open);

using var mediaPipe = new MediaPipeDetector(
    options: MediaPipeOptions.FromStream(stream)
);

Sources

1. Original MediaPipe Repository
2. Language detection guide
3. Language detector (model card)
4. Language identification at Wikipedia

We value your feedback. Feel free to open issues or contribute to the repository. Let’s make language detection in .NET even more powerful and versatile! 🌍📝

Happy coding! 👩‍💻👨‍💻