Machine learning service helps in application text language online and offline detection

When users need cross language communication or content translation , The application needs to be able to automatically detect the language of the text before translation .
HMS Core The language detection service of machine learning service provides online language detection and offline language detection , It supports the detection of monolingual text , It also supports detecting mixed language text , Covers Afrikaans 、 Arabic and other hundreds of languages . Access language detection service ,App Translation language detection can be easily realized 、 Web page language detection , Language detection in mixed language scenarios , help App Enhance user experience .
Language detection process
Input text , Machine learning service language automatically detects the language of the text , Then return the corresponding language code and the corresponding confidence , Or return an application encoded in the language with the highest confidence to the developer . Finally under development app The page presents the translated text .

DSC0000.png(188.63 KB, amount of downloads : 0)

Download the attachment

2  Hours before Upload

Effect demonstration

DSC0001.gif(1.67 MB, amount of downloads : 0)

Download the attachment

2  Hours before Upload

1. The development of preparation
Before development, you need to configure HMS Core SDK Of Maven Warehouse address

  
 
 

   
  
  
repositories {
   
  
  
 maven { 
   
  
  
url'https://cmc.centralrepo.rnd.huawei.com/artifactory/product_maven/' }
   
  
  
}
  
 
 

Integrated online language detection service SDK, The sample code is as follows

  
 
 

   
  
  
dependencies{
   
  
  
 implementation 'com.huawei.hms:ml-computer-language-detection:3.4.0.301'
   
  
  
}
  
 
 

2. Edit project integration
2.1 Set the authentication information of the application

  
 
 

   
  
  
 Can pass api_key perhaps Access Token To set the application authentication information .
   
  
  
 adopt setAccessToken Method setting Access Token, You can initialize the settings once when the application starts , There is no need to set it more than once .
   
  
  
MLApplication.getInstance().setAccessToken("your access token");
   
  
  
 adopt setApiKey Method setting api_key, You can initialize the settings once when the application starts , There is no need to set it more than once .
   
  
  
MLApplication.getInstance().setApiKey("your ApiKey");
  
 
 

2.2 Create a language detector

  
 
 

   
  
  
//  Mode one ： Create a language detector using the default parameter configuration .
   
  
  
MLRemoteLangDetector mlRemoteLangDetector = MLLangDetectorFactory.getInstance()
   
  
  
 .getRemoteLangDetector();
   
  
  
//  Mode two ： Create a language detector using a custom parameter configuration .
   
  
  
MLRemoteLangDetectorSetting setting = new MLRemoteLangDetectorSetting.Factory()
   
  
  
 //  Set the minimum confidence threshold for language detection .
   
  
  
 .setTrustedThreshold(0.01f)
   
  
  
 .create();
   
  
  
MLRemoteLangDetector mlRemoteLangDetector = MLLangDetectorFactory.getInstance()
   
  
  
 .getRemoteLangDetector(setting);
  
 
 

2.3 Conduct language detection
Asynchronous method sample code

  
 
 

   
  
  
//  Mode one ： Returns the detection results of multiple languages , Including language coding and confidence ,sourceText Represents the text to be detected , Length should be less than 5000 Characters .
   
  
  
Task<List<MLDetectedLang>> probabilityDetectTask = mlRemoteLangDetector.probabilityDetect(sourceText);
   
  
  
probabilityDetectTask.addOnSuccessListener(new OnSuccessListener<List<MLDetectedLang>>() {
   
  
  
 @Override
   
  
  
 public void onSuccess(List<MLDetectedLang> result) {
   
  
  
 //  Successful processing logic .
   
  
  
 }
   
  
  
}).addOnFailureListener(new OnFailureListener() {
   
  
  
 @Override
   
  
  
 public void onFailure(Exception e) {
   
  
  
 //  Failed processing logic .
   
  
  
 // Recognition failure.
   
  
  
 try {
   
  
  
 MLException mlException = (MLException)e;
   
  
  
 //  Get error code , Developers can handle error codes , According to the error code for differentiated page prompt .
   
  
  
 int errorCode = mlException.getErrCode();
   
  
  
 //  Get error information , Developers can combine error codes , Fast location problem .
   
  
  
 String errorMessage = mlException.getMessage();
   
  
  
 } catch (Exception error) {
   
  
  
 //  Conversion error handling .
   
  
  
 }
   
  
  
 }
   
  
  
});
   
  
  
//  Mode two ： Returns the language code with the highest confidence ,sourceText Represents the text to be detected , Length should be less than 5000 Characters .
   
  
  
Task<String> firstBestDetectTask = mlRemoteLangDetector.firstBestDetect(sourceText);
   
  
  
firstBestDetectTask.addOnSuccessListener(new OnSuccessListener<String>() {
   
  
  
 @Override
   
  
  
 public void onSuccess(String s) {
   
  
  
 //  Successful processing logic .
   
  
  
 }
   
  
  
}).addOnFailureListener(new OnFailureListener() {
   
  
  
 @Override
   
  
  
 public void onFailure(Exception e) {
   
  
  
 //  Failed processing logic .
   
  
  
 // Recognition failure.
   
  
  
 try {
   
  
  
 MLException mlException = (MLException)e;
   
  
  
 //  Get error code , Developers can handle error codes , According to the error code for differentiated page prompt .
   
  
  
 int errorCode = mlException.getErrCode();
   
  
  
 //  Get error information , Developers can combine error codes , Fast location problem .
   
  
  
 String errorMessage = mlException.getMessage();
   
  
  
 } catch (Exception error) {
   
  
  
 //  Conversion error handling .
   
  
  
 }
   
  
  
 }
   
  
  
});
  
 
 

Synchronization method sample code

  
 
 

   
  
  
//  Mode one ： Returns the detection results of multiple languages , Including language coding and confidence ,sourceText Represents the text to be detected , Length should be less than 5000 Characters .
   
  
  
try {
   
  
  
 List<MLDetectedLang> result= mlRemoteLangDetector.syncProbabilityDetect(sourceText);
   
  
  
} catch (MLException mlException) {
   
  
  
 //  Failed processing logic .
   
  
  
 //  Get error code , Developers can handle error codes , According to the error code for differentiated page prompt .
   
  
  
 int errorCode = mlException.getErrCode();
   
  
  
 //  Get error information , Developers can combine error codes , Fast location problem .
   
  
  
 String errorMessage = mlException.getMessage();
   
  
  
}
   
  
  
//  Mode two ： Returns the language code with the highest confidence ,sourceText Represents the text to be detected , Length should be less than 5000 Characters .
   
  
  
try {
   
  
  
 String language = mlRemoteLangDetector.syncFirstBestDetect(sourceText);
   
  
  
} catch (MLException mlException) {
   
  
  
 //  Failed processing logic .
   
  
  
 //  Get error code , Developers can handle error codes , According to the error code for differentiated page prompt .
   
  
  
 int errorCode = mlException.getErrCode();
   
  
  
 //  Get error information , Developers can combine error codes , Fast location problem .
   
  
  
 String errorMessage = mlException.getMessage();
   
  
  
}
  
 
 

2.4 After the test , Release resources .

  
 
 

   
  
  
if (mlRemoteLangDetector != null) {
   
  
  
 mlRemoteLangDetector.stop();
   
  
  
}
  
 
 

The language detection function includes the end side and cloud side , The above examples are only cloud side functions .
Learn more >>
Visit the official website of Huawei developer alliance
Get development guidance document
Huawei mobile service open source warehouse address ：GitHub、Gitee
Pay attention to our , The first time to understand HMS Core Latest technical information ~