在 iOS 上使用 ML Kit 翻譯文字

你可以使用 ML Kit 翻譯不同語言的文字。機器學習套件 目前支援 59 種語言

事前準備

  1. 如果尚未將 Firebase 加入應用程式,請按照下列步驟操作: 入門指南中的步驟。
  2. 在 Podfile 中加入 ML Kit 程式庫:
    pod 'Firebase/MLNLTranslate', '6.25.0'
    
    敬上 安裝或更新專案的 Pod 後,請務必開啟 Xcode 專案 .xcworkspace
  3. 在應用程式中匯入 Firebase:
    import Firebase
    @import Firebase;

翻譯一段文字

如何在兩種語言之間翻譯字串:

  1. 建立 Translator 物件,並使用來源和目標設定該物件 語言:

    // Create an English-German translator:
    let options = TranslatorOptions(sourceLanguage: .en, targetLanguage: .de)
    let englishGermanTranslator = NaturalLanguage.naturalLanguage().translator(options: options)
    
    // Create an English-German translator:
    FIRTranslatorOptions *options =
        [[FIRTranslatorOptions alloc] initWithSourceLanguage:FIRTranslateLanguageEN
                                              targetLanguage:FIRTranslateLanguageDE];
    FIRTranslator *englishGermanTranslator =
        [[FIRNaturalLanguage naturalLanguage] translatorWithOptions:options];
    

    如果您不知道輸入文字的語言,可以使用 識別資訊 API。(但別忘了 而不是同時在裝置上保留太多語言模型)。

  2. 確認所需翻譯模型已下載到裝置。 確認模型可用前,請勿呼叫 translate(_:completion:)

    let conditions = ModelDownloadConditions(
        allowsCellularAccess: false,
        allowsBackgroundDownloading: true
    )
    englishGermanTranslator.downloadModelIfNeeded(with: conditions) { error in
        guard error == nil else { return }
    
        // Model downloaded successfully. Okay to start translating.
    }
    
    FIRModelDownloadConditions *conditions =
        [[FIRModelDownloadConditions alloc] initWithAllowsCellularAccess:NO
                                             allowsBackgroundDownloading:YES];
    [englishGermanTranslator downloadModelIfNeededWithConditions:conditions
                                                      completion:^(NSError *_Nullable error) {
      if (error != nil) {
        return;
      }
      // Model downloaded successfully. Okay to start translating.
    }];
    

    語言模型大約有 30 MB 的空間,因此不需要下載 除非使用者另有指定,否則只能透過 Wi-Fi 下載下載。個人中心 並刪除不需要的模型 請參閱「明確管理翻譯模型」。

  3. 確認下載模型後,請在 將原文語言轉換為 translate(_:completion:)

    englishGermanTranslator.translate(text) { translatedText, error in
        guard error == nil, let translatedText = translatedText else { return }
    
        // Translation succeeded.
    }
    
    [englishGermanTranslator translateText:text
                                completion:^(NSString *_Nullable translatedText,
                                             NSError *_Nullable error) {
      if (error != nil || translatedText == nil) {
        return;
      }
    
      // Translation succeeded.
    }];
    

    ML Kit 會將文字翻譯成您設定的目標語言 將翻譯的文字傳遞至完成處理常式。

明確管理翻譯模型

按照上述方式使用翻譯 API 時,ML Kit 會自動使用 視需要下載特定語言專屬的翻譯模型至裝置。個人中心 也可以明確管理 整合了 ML Kit 的翻譯模型管理 API可用的值包括 舉例來說,如果您想事先下載模型,或是刪除不需要的模型 登入裝置。

如要取得儲存在裝置上的翻譯模型:

let localModels = ModelManager.modelManager().downloadedTranslateModels
NSSet<FIRTranslateRemoteModel *> *localModels =
    [FIRModelManager modelManager].downloadedTranslateModels;

如要刪除模型:

// Delete the German model if it's on the device.
let deModel = TranslateRemoteModel.translateRemoteModel(language: .de)
ModelManager.modelManager().deleteDownloadedModel(deModel) { error in
    guard error == nil else { return }
    // Model deleted.
}
// Delete the German model if it's on the device.
FIRTranslateRemoteModel *deModel =
    [FIRTranslateRemoteModel translateRemoteModelWithLanguage:FIRTranslateLanguageDE];
[[FIRModelManager modelManager] deleteDownloadedModel:deModel
                                           completion:^(NSError * _Nullable error) {
                                               if (error != nil) {
                                                   return;
                                               }
                                               // Model deleted.
                                           }];

如何下載模型:

// Download the French model.
let frModel = TranslateRemoteModel.translateRemoteModel(language: .fr)

// Keep a reference to the download progress so you can check that the model
// is available before you use it.
progress = ModelManager.modelManager().download(
    frModel,
    conditions: ModelDownloadConditions(
        allowsCellularAccess: false,
        allowsBackgroundDownloading: true
    )
)

如要向「NotificationCenter」取得下載狀態,請註冊 firebaseMLModelDownloadDidSucceedfirebaseMLModelDownloadDidFail。請務必對「self」使用較弱的參照 ,因為下載可能需要一些時間,而且 物件即可釋出。例如:

NotificationCenter.default.addObserver(
    forName: .firebaseMLModelDownloadDidSucceed,
    object: nil,
    queue: nil
) { [weak self] notification in
    guard let strongSelf = self,
        let userInfo = notification.userInfo,
        let model = userInfo[ModelDownloadUserInfoKey.remoteModel.rawValue]
            as? TranslateRemoteModel,
        model == frModel
        else { return }
    // The model was downloaded and is available on the device
}

NotificationCenter.default.addObserver(
    forName: .firebaseMLModelDownloadDidFail,
    object: nil,
    queue: nil
) { [weak self] notification in
    guard let strongSelf = self,
        let userInfo = notification.userInfo,
        let model = userInfo[ModelDownloadUserInfoKey.remoteModel.rawValue]
            as? TranslateRemoteModel
        else { return }
    let error = userInfo[ModelDownloadUserInfoKey.error.rawValue]
    // ...
}
// Download the French model.
FIRModelDownloadConditions *conditions =
    [[FIRModelDownloadConditions alloc] initWithAllowsCellularAccess:NO
                                         allowsBackgroundDownloading:YES];
FIRTranslateRemoteModel *frModel =
    [FIRTranslateRemoteModel translateRemoteModelWithLanguage:FIRTranslateLanguageFR];

// Keep a reference to the download progress so you can check that the model
// is available before you use it.
self.downloadProgress = [[FIRModelManager modelManager] downloadModel:frModel
                                                           conditions:conditions];

如要向「NSNotificationCenter」取得下載狀態,請註冊 FIRModelDownloadDidSucceedNotificationFIRModelDownloadDidFailNotification。請務必使用弱式參照 觀察器區塊中的 self,因為下載可能需要一些時間, 可在下載完成後釋放。

__block MyViewController *weakSelf = self;

[NSNotificationCenter.defaultCenter
 addObserverForName:FIRModelDownloadDidSucceedNotification
 object:nil
 queue:nil
 usingBlock:^(NSNotification * _Nonnull note) {
     if (weakSelf == nil | note.userInfo == nil) {
         return;
     }

     FIRTranslateRemoteModel *model = note.userInfo[FIRModelDownloadUserInfoKeyRemoteModel];
     if ([model isKindOfClass:[FIRTranslateRemoteModel class]]
         && model == frModel) {
         // The model was downloaded and is available on the device
     }
 }];

[NSNotificationCenter.defaultCenter
 addObserverForName:FIRModelDownloadDidFailNotification
 object:nil
 queue:nil
 usingBlock:^(NSNotification * _Nonnull note) {
     if (weakSelf == nil | note.userInfo == nil) {
         return;
     }

     NSError *error = note.userInfo[FIRModelDownloadUserInfoKeyError];
 }];

ML Kit for Firebase provided ready-to-use ML solutions for app developers. New apps should use the standalone ML Kit library for on-device ML and Firebase ML for cloud-based ML.

更新時間:Oct 23, 2024