Create aTranslatorobject, configuring it with the source and target languages:
Kotlin
// Create an English-German translator:valoptions=TranslatorOptions.Builder().setSourceLanguage(TranslateLanguage.ENGLISH).setTargetLanguage(TranslateLanguage.GERMAN).build()valenglishGermanTranslator=Translation.getClient(options)
Java
// Create an English-German translator:TranslatorOptionsoptions=newTranslatorOptions.Builder().setSourceLanguage(TranslateLanguage.ENGLISH).setTargetLanguage(TranslateLanguage.GERMAN).build();finalTranslatorenglishGermanTranslator=Translation.getClient(options);
Avoid keeping too many language models on the device at once.
Make sure the required translation model has been downloaded to the device.
Don't calltranslate()until you know the model is available.
Kotlin
varconditions=DownloadConditions.Builder().requireWifi().build()englishGermanTranslator.downloadModelIfNeeded(conditions).addOnSuccessListener{// Model downloaded successfully. Okay to start translating.// (Set a flag, unhide the translation UI, etc.)}.addOnFailureListener{exception->// Model couldn’t be downloaded or other internal error.// ...}
Java
DownloadConditionsconditions=newDownloadConditions.Builder().requireWifi().build();englishGermanTranslator.downloadModelIfNeeded(conditions).addOnSuccessListener(newOnSuccessListener(){@OverridepublicvoidonSuccess(Voidv){// Model downloaded successfully. Okay to start translating.// (Set a flag, unhide the translation UI, etc.)}}).addOnFailureListener(newOnFailureListener(){@OverridepublicvoidonFailure(@NonNullExceptione){// Model couldn’t be downloaded or other internal error.// ...}});
Language models are around 30MB, so don't download them unnecessarily, and
only download them using Wi-Fi unless the user has specified otherwise. You
should also delete unneeded models.
SeeExplicitly manage translation models.
After you confirm the model has been downloaded, pass a string of text in
the source language totranslate():
The translated text, in the target language you configured, is passed to the
success listener.
Ensure that theclose()method is called when theTranslatorobject will no longer be used.
If you are using a Translator in a Fragment or AppCompatActivity, one easy
way to do that is callLifecycleOwner.getLifecycle()on the Fragment or AppCompatActivity, and then callLifecycle.addObserver.
For example:
Note that this assumes that the code is inside of a class that implements
LifecycleOwner (e.g. a Fragment or AppCompatActivity).
Explicitly manage translation models
When you use the translation API as described above, ML Kit automatically
downloads language-specific translation models to the device as required. You
can also explicitly manage the translation models you want available on the
device by using ML Kit's translation model management API. This can be
useful if you want to download models ahead of time, or delete unneeded models
from the device.
Kotlin
valmodelManager=RemoteModelManager.getInstance()// Get translation models stored on the device.modelManager.getDownloadedModels(TranslateRemoteModel::class.java).addOnSuccessListener{models->// ...}.addOnFailureListener{// Error.}// Delete the German model if it's on the device.valgermanModel=TranslateRemoteModel.Builder(TranslateLanguage.GERMAN).build()modelManager.deleteDownloadedModel(germanModel).addOnSuccessListener{// Model deleted.}.addOnFailureListener{// Error.}// Download the French model.valfrenchModel=TranslateRemoteModel.Builder(TranslateLanguage.FRENCH).build()valconditions=DownloadConditions.Builder().requireWifi().build()modelManager.download(frenchModel,conditions).addOnSuccessListener{// Model downloaded.}.addOnFailureListener{// Error.}
Java
RemoteModelManagermodelManager=RemoteModelManager.getInstance();// Get translation models stored on the device.modelManager.getDownloadedModels(TranslateRemoteModel.class).addOnSuccessListener(newOnSuccessListener<Set>(){@OverridepublicvoidonSuccess(Setmodels){// ...}}).addOnFailureListener(newOnFailureListener(){@OverridepublicvoidonFailure(@NonNullExceptione){// Error.}});// Delete the German model if it's on the device.TranslateRemoteModelgermanModel=newTranslateRemoteModel.Builder(TranslateLanguage.GERMAN).build();modelManager.deleteDownloadedModel(germanModel).addOnSuccessListener(newOnSuccessListener(){@OverridepublicvoidonSuccess(Voidv){// Model deleted.}}).addOnFailureListener(newOnFailureListener(){@OverridepublicvoidonFailure(@NonNullExceptione){// Error.}});// Download the French model.TranslateRemoteModelfrenchModel=newTranslateRemoteModel.Builder(TranslateLanguage.FRENCH).build();DownloadConditionsconditions=newDownloadConditions.Builder().requireWifi().build();modelManager.download(frenchModel,conditions).addOnSuccessListener(newOnSuccessListener(){@OverridepublicvoidonSuccess(Voidv){// Model downloaded.}}).addOnFailureListener(newOnFailureListener(){@OverridepublicvoidonFailure(@NonNullExceptione){// Error.}});
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[[["\u003cp\u003eML Kit can translate text between more than 50 languages using on-demand dynamic model downloads.\u003c/p\u003e\n"],["\u003cp\u003eYou need to create a \u003ccode\u003eTranslator\u003c/code\u003e object, download the necessary language model, and then use the \u003ccode\u003etranslate()\u003c/code\u003e method to translate text.\u003c/p\u003e\n"],["\u003cp\u003eExplicit model management allows you to pre-download, delete, and control language models on the device for optimized performance.\u003c/p\u003e\n"],["\u003cp\u003eIt's recommended to download models over Wi-Fi and delete unused models to save storage space and bandwidth.\u003c/p\u003e\n"],["\u003cp\u003eEnsure the \u003ccode\u003eclose()\u003c/code\u003e method is called when the \u003ccode\u003eTranslator\u003c/code\u003e object is no longer needed, potentially using LifecycleObserver for easier management in Fragments or Activities.\u003c/p\u003e\n"]]],["ML Kit facilitates text translation between over 50 languages. To use it, add the necessary dependencies to your project. Create a `Translator` object, specifying source and target languages. Download the required language model using `downloadModelIfNeeded`, preferably over Wi-Fi. After confirming the download, translate text using the `translate()` method. Finally, call the `close()` method when finished. Models can also be managed explicitly to download or delete them.\n"],null,["You can use ML Kit to translate text between languages. ML Kit\ncan translation between more than\n[50 languages](/ml-kit/languages/translation/translation-language-support).\n\n\u003cbr /\u003e\n\n| This API supports on demand dynamic model downloads. See [this guide](/ml-kit/tips/installation-paths) for more information.\n\nTry it out\n\n- Play around with [the sample app](https://github.com/googlesamples/mlkit/tree/master/android/translate) to see an example usage of this API.\n\nBefore you begin This API requires Android API level 21 or above. Make sure that your app's build file uses a `minSdkVersion` value of 21 or higher.\n\n1. In your project-level `build.gradle` file, make sure to include Google's Maven repository in both your `buildscript` and `allprojects` sections.\n2. Add the dependencies for the ML Kit Android libraries to your module's app-level gradle file, which is usually `app/build.gradle`: \n\n ```carbon\n dependencies {\n // ...\n\n implementation 'com.google.mlkit:translate:17.0.3'\n }\n ```\n\nTranslate a string of text\n\nTo translate a string between two languages:\n\n1. Create a\n [`Translator`](/android/reference/com/google/mlkit/nl/translate/Translator)\n object, configuring it with the source and target languages:\n\n Kotlin \n\n ```kotlin\n // Create an English-German translator:\n val options = TranslatorOptions.Builder()\n .setSourceLanguage(TranslateLanguage.ENGLISH)\n .setTargetLanguage(TranslateLanguage.GERMAN)\n .build()\n val englishGermanTranslator = Translation.getClient(options)\n ```\n\n Java \n\n ```java\n // Create an English-German translator:\n TranslatorOptions options =\n new TranslatorOptions.Builder()\n .setSourceLanguage(TranslateLanguage.ENGLISH)\n .setTargetLanguage(TranslateLanguage.GERMAN)\n .build();\n final Translator englishGermanTranslator =\n Translation.getClient(options);\n ```\n\n If you don't know the language of the input text, you can use the [Language\n Identification API](/ml-kit/language/identification) which\n gives you a language tag. Then convert the tag to a `TranslateLanguage`\n using [`TranslateLanguage.fromLanguageTag()`](/android/reference/com/google/mlkit/nl/translate/TranslateLanguage#public-static-string-fromlanguagetag-string-languagetag).\n\n Avoid keeping too many language models on the device at once.\n2. Make sure the required translation model has been downloaded to the device.\n Don't call `translate()` until you know the model is available.\n\n Kotlin \n\n ```kotlin\n var conditions = DownloadConditions.Builder()\n .requireWifi()\n .build()\n englishGermanTranslator.downloadModelIfNeeded(conditions)\n .addOnSuccessListener {\n // Model downloaded successfully. Okay to start translating.\n // (Set a flag, unhide the translation UI, etc.)\n }\n .addOnFailureListener { exception -\u003e\n // Model couldn't be downloaded or other internal error.\n // ...\n }\n ```\n\n Java \n\n ```java\n DownloadConditions conditions = new DownloadConditions.Builder()\n .requireWifi()\n .build();\n englishGermanTranslator.downloadModelIfNeeded(conditions)\n .addOnSuccessListener(\n new OnSuccessListener() {\n @Override\n public void onSuccess(Void v) {\n // Model downloaded successfully. Okay to start translating.\n // (Set a flag, unhide the translation UI, etc.)\n }\n })\n .addOnFailureListener(\n new OnFailureListener() {\n @Override\n public void onFailure(@NonNull Exception e) {\n // Model couldn't be downloaded or other internal error.\n // ...\n }\n });\n ```\n\n Language models are around 30MB, so don't download them unnecessarily, and\n only download them using Wi-Fi unless the user has specified otherwise. You\n should also delete unneeded models.\n See [Explicitly manage translation models](#manage-models).\n3. After you confirm the model has been downloaded, pass a string of text in\n the source language to\n [`translate()`](/android/reference/com/google/mlkit/nl/translate/Translator#public-taskstring-translate-string-input):\n\n Kotlin \n\n ```kotlin\n englishGermanTranslator.translate(text)\n .addOnSuccessListener { translatedText -\u003e\n // Translation successful.\n }\n .addOnFailureListener { exception -\u003e\n // Error.\n // ...\n }\n ```\n\n Java \n\n ```java\n englishGermanTranslator.translate(text)\n .addOnSuccessListener(\n new OnSuccessListener() {\n @Override\n public void onSuccess(@NonNull String translatedText) {\n // Translation successful.\n }\n })\n .addOnFailureListener(\n new OnFailureListener() {\n @Override\n public void onFailure(@NonNull Exception e) {\n // Error.\n // ...\n }\n });\n ```\n\n The translated text, in the target language you configured, is passed to the\n success listener.\n4. Ensure that the [`close()`](/android/reference/com/google/mlkit/nl/translate/Translator#public-void-close)\n method is called when the\n [`Translator`](/android/reference/com/google/mlkit/nl/translate/Translator)\n object will no longer be used.\n\n If you are using a Translator in a Fragment or AppCompatActivity, one easy\n way to do that is call [LifecycleOwner.getLifecycle()](https://developer.android.com/reference/androidx/lifecycle/LifecycleOwner#getLifecycle())\n on the Fragment or AppCompatActivity, and then call [Lifecycle.addObserver](https://developer.android.com/reference/androidx/lifecycle/Lifecycle#addObserver(androidx.lifecycle.LifecycleObserver)).\n For example: \n\n Kotlin \n\n ```kotlin\n val options = ...\n val translator = Translation.getClient(options)\n getLifecycle().addObserver(translator)\n ```\n\n Java \n\n ```java\n TranslatorOptions options = ...\n Translator translator = Translation.getClient(options);\n getLifecycle().addObserver(translator);\n ... use translator ...\n ```\n\n Note that this assumes that the code is inside of a class that implements\n LifecycleOwner (e.g. a Fragment or AppCompatActivity).\n\nExplicitly manage translation models\n\n\nWhen you use the translation API as described above, ML Kit automatically\ndownloads language-specific translation models to the device as required. You\ncan also explicitly manage the translation models you want available on the\ndevice by using ML Kit's translation model management API. This can be\nuseful if you want to download models ahead of time, or delete unneeded models\nfrom the device.\n\n\u003cbr /\u003e\n\nKotlin \n\n```kotlin\nval modelManager = RemoteModelManager.getInstance()\n\n// Get translation models stored on the device.\nmodelManager.getDownloadedModels(TranslateRemoteModel::class.java)\n .addOnSuccessListener { models -\u003e\n // ...\n }\n .addOnFailureListener {\n // Error.\n }\n\n// Delete the German model if it's on the device.\nval germanModel = TranslateRemoteModel.Builder(TranslateLanguage.GERMAN).build()\nmodelManager.deleteDownloadedModel(germanModel)\n .addOnSuccessListener {\n // Model deleted.\n }\n .addOnFailureListener {\n // Error.\n }\n\n// Download the French model.\nval frenchModel = TranslateRemoteModel.Builder(TranslateLanguage.FRENCH).build()\nval conditions = DownloadConditions.Builder()\n .requireWifi()\n .build()\nmodelManager.download(frenchModel, conditions)\n .addOnSuccessListener {\n // Model downloaded.\n }\n .addOnFailureListener {\n // Error.\n }\n```\n\nJava \n\n```java\nRemoteModelManager modelManager = RemoteModelManager.getInstance();\n\n// Get translation models stored on the device.\nmodelManager.getDownloadedModels(TranslateRemoteModel.class)\n .addOnSuccessListener(new OnSuccessListener\u003cSet\u003e() {\n @Override\n public void onSuccess(Set models) {\n // ...\n }\n })\n .addOnFailureListener(new OnFailureListener() {\n @Override\n public void onFailure(@NonNull Exception e) {\n // Error.\n }\n });\n\n// Delete the German model if it's on the device.\nTranslateRemoteModel germanModel =\n new TranslateRemoteModel.Builder(TranslateLanguage.GERMAN).build();\nmodelManager.deleteDownloadedModel(germanModel)\n .addOnSuccessListener(new OnSuccessListener() {\n @Override\n public void onSuccess(Void v) {\n // Model deleted.\n }\n })\n .addOnFailureListener(new OnFailureListener() {\n @Override\n public void onFailure(@NonNull Exception e) {\n // Error.\n }\n });\n\n// Download the French model.\nTranslateRemoteModel frenchModel =\n new TranslateRemoteModel.Builder(TranslateLanguage.FRENCH).build();\nDownloadConditions conditions = new DownloadConditions.Builder()\n .requireWifi()\n .build();\nmodelManager.download(frenchModel, conditions)\n .addOnSuccessListener(new OnSuccessListener() {\n @Override\n public void onSuccess(Void v) {\n // Model downloaded.\n }\n })\n .addOnFailureListener(new OnFailureListener() {\n @Override\n public void onFailure(@NonNull Exception e) {\n // Error.\n }\n });\n```"]]
