How to get translation groups from Microsoft Translator?

If you go to https://www.bing.com/translator, (which uses the MS/Azure Translator api) and type in the word mean from English to Swedish, in addition to the “main” translation you get on the right, you also have a section that has “Other ways to say”, which are grouped by Verb, Noun, and Adjective.

I would like to know how I can fetch this list of groups from the response.

Right now I have the following, but it only returns the main translation, in this case Menar.

import com.fasterxml.jackson.databind.ObjectMapper;
import com.squareup.okhttp.MediaType;
import com.squareup.okhttp.OkHttpClient;
import com.squareup.okhttp.Protocol;
import com.squareup.okhttp.Request;
import com.squareup.okhttp.RequestBody;
import com.squareup.okhttp.Response;

protected String doInBackground(String... params) {

    String word = params[0];
    String translationType = params[1];

    MediaType mediaType = MediaType.parse("application/json");
    RequestBody body = RequestBody.create(mediaType,
            "[{nt"Text": "" + word + ""n}]");
    Request request = new Request.Builder()
            .url(BASE_URL + translationType)
            .post(body)
            .addHeader("Ocp-Apim-Subscription-Key", SUBSCRIPTION_KEY)
            .addHeader("Ocp-Apim-Subscription-Region", SUBSCRIPTION_REGION)
            .addHeader("Content-type", "application/json")
            .build();

    Response response = okHttpClient.newCall(request)
            .execute();
    if (!response.isSuccessful()) {
        throw new AzureTranslateException("Failed to get translations from Azure Translator API, due to: "
                + response.message());
    }
    String json = response.body().string();
    // remove the first and last characters, which are brackets, for ObjectMapper
    json = json.substring(1, json.length() - 1);

    // this will only have ONE translation
    AzureTranslateResponse r = new ObjectMapper().readValue(json, AzureTranslateResponse.class);

    return r.getTranslations().get(0).getText();
}

AzureTranslatorResponse

@Data
public class AzureTranslateResponse {

    private DetectedLanguage detectedLanguage;
    private List<Translation> translations;
}

DetectedLanguage

@Data
public class DetectedLanguage {

    private String language;
    private double score;
}

DetectedLanguage

@Data
public class DetectedLanguage {

    private String language;
    private double score;
}

Answer

You can retrieve alternative translations using the Dictionary Lookup resource. https://learn.microsoft.com/azure/cognitive-services/translator/reference/v3-0-dictionary-lookup

It returns the part of speech in the posTag attribute. You can then group by posTag to achieve a similar grouping.

The Dictionary Examples resource returns the example sentences that you see on the Bing Translator site as well. https://learn.microsoft.com/azure/cognitive-services/translator/reference/v3-0-dictionary-examples

Advertisement

Answer