安装或更新项目的 Pod 之后,使用 Xcode 项目的
.xcworkspace。Xcode 12.4 或更高版本支持机器学习套件。
1. 创建对话历史记录对象
要生成智能回复,您需要向机器学习套件传递
TextMessage 对象,时间戳早的在前面。每当用户
发送或接收消息,添加消息、其时间戳和消息
将发送者的用户 ID 添加到对话记录中。
用户 ID 可以是任何字符串,用于标识
对话。User ID 不需要与任何用户数据相对应
并且用户 ID 无需在会话之间保持一致,
调用智能回复生成器。
如果消息是由您要提供回复建议的用户发送的,则将
isLocalUser 设为 true。
Swift
varconversation:[TextMessage]=[]// Then, for each message sent and received:letmessage=TextMessage(text:"How are you?",timestamp:Date().timeIntervalSince1970,userID:"userId",isLocalUser:false)conversation.append(message)
Objective-C
NSMutableArray*conversation=[NSMutableArrayarray];// Then, for each message sent and received:MLKTextMessage*message=[[MLKTextMessagealloc]initWithText:@"How are you?"timestamp:[NSDatedate].timeIntervalSince1970userID:userIdisLocalUser:NO];[conversationaddObject:message];
SmartReply.smartReply().suggestReplies(for:conversation){result,erroringuarderror==nil,letresult=resultelse{return}if(result.status==.notSupportedLanguage){// The conversation's language isn't supported, so// the result doesn't contain any suggestions.}elseif(result.status==.success){// Successfully suggested smart replies.// ...}}
Objective-C
MLKSmartReply*smartReply=[MLKSmartReplysmartReply];[smartReplysuggestRepliesForMessages:inputTextcompletion:^(MLKSmartReplySuggestionResult*_Nullableresult,NSError*_Nullableerror){if(error||!result){return;}if(result.status==MLKSmartReplyResultStatusNotSupportedLanguage){// The conversation's language isn't supported, so// the result doesn't contain any suggestions.}elseif(result.status==MLKSmartReplyResultStatusSuccess){// Successfully suggested smart replies.// ...}}];
[null,null,["最后更新时间 (UTC):2025-08-29。"],[[["\u003cp\u003eML Kit provides an on-device model to generate smart replies for messages in English conversations, enhancing user experience and engagement.\u003c/p\u003e\n"],["\u003cp\u003eBy passing a conversation history to ML Kit, developers can receive up to three suggested replies for the latest message, which can then be displayed to the user.\u003c/p\u003e\n"],["\u003cp\u003eBefore utilizing the API, ensure the device is 64-bit and include the necessary ML Kit pods in your project.\u003c/p\u003e\n"],["\u003cp\u003eThe smart reply feature is optimized for non-sensitive conversations, and may not generate results if the language is unsupported or sensitive topics are detected.\u003c/p\u003e\n"]]],[],null,["ML Kit can generate short replies to messages using an on-device model.\n\nTo generate smart replies, you pass ML Kit a log of recent messages in a\nconversation. If ML Kit determines the conversation is in English, and that\nthe conversation doesn't have potentially sensitive subject matter, ML Kit\ngenerates up to three replies, which you can suggest to your user.\n\n\u003cbr /\u003e\n\n| **Note:** ML Kit iOS APIs only run on 64-bit devices. If you build your app with 32-bit support, check the device's architecture before using this API.\n\nTry it out\n\n- Play around with [the sample app](https://github.com/googlesamples/mlkit/tree/master/ios/quickstarts/smartreply) to see an example usage of this API.\n\nBefore you begin\n\n1. Include the following ML Kit pods in your Podfile: \n\n ```\n pod 'GoogleMLKit/SmartReply', '8.0.0'\n ```\n2. After you install or update your project's Pods, open your Xcode project using its `.xcworkspace`. ML Kit is supported in Xcode version 12.4 or greater.\n\n1. Create a conversation history object\n\nTo generate smart replies, you pass ML Kit a chronologically-ordered array of\n`TextMessage` objects, with the earliest timestamp first. Whenever the user\nsends or receives a message, add the message, its timestamp, and the message\nsender's user ID to the conversation history.\n\nThe user ID can be any string that uniquely identifies the sender within the\nconversation. The user ID doesn't need to correspond to any user data,\nand the user ID doesn't need to be consistent between conversations or\ninvocations of the smart reply generator.\n\nIf the message was sent by the user you want to suggest replies to, set\n`isLocalUser` to true. \n\nSwift \n\n```swift\nvar conversation: [TextMessage] = []\n\n// Then, for each message sent and received:\nlet message = TextMessage(\n text: \"How are you?\",\n timestamp: Date().timeIntervalSince1970,\n userID: \"userId\",\n isLocalUser: false)\nconversation.append(message)\n```\n\nObjective-C \n\n```objective-c\nNSMutableArray *conversation = [NSMutableArray array];\n\n// Then, for each message sent and received:\nMLKTextMessage *message = [[MLKTextMessage alloc]\n initWithText:@\"How are you?\"\n timestamp:[NSDate date].timeIntervalSince1970\n userID:userId\n isLocalUser:NO];\n[conversation addObject:message];\n```\n\nA conversation history object looks like the following example:\n\n| Timestamp | userID | isLocalUser | Message |\n|------------------------------|---------|-------------|----------------------|\n| Thu Feb 21 13:13:39 PST 2019 | | true | are you on your way? |\n| Thu Feb 21 13:15:03 PST 2019 | FRIEND0 | false | Running late, sorry! |\n\nML Kit suggests replies to the last message in a conversation history. The last message\nshould be from a non-local user. In the example above, the last message in the conversation\nis from the non-local user FRIEND0. When you use pass ML Kit this log, it suggests\nreplies to FRIENDO's message: \"Running late, sorry!\"\n\n2. Get message replies\n\nTo generate smart replies to a message, get an instance of `SmartReply` and pass\nthe conversation history to its `suggestReplies(for:completion:)` method: \n\nSwift \n\n```swift\nSmartReply.smartReply().suggestReplies(for: conversation) { result, error in\n guard error == nil, let result = result else {\n return\n }\n if (result.status == .notSupportedLanguage) {\n // The conversation's language isn't supported, so\n // the result doesn't contain any suggestions.\n } else if (result.status == .success) {\n // Successfully suggested smart replies.\n // ...\n }\n}\n```\n\nObjective-C \n\n```objective-c\nMLKSmartReply *smartReply = [MLKSmartReply smartReply];\n[smartReply suggestRepliesForMessages:inputText\n completion:^(MLKSmartReplySuggestionResult * _Nullable result,\n NSError * _Nullable error) {\n if (error || !result) {\n return;\n }\n if (result.status == MLKSmartReplyResultStatusNotSupportedLanguage) {\n // The conversation's language isn't supported, so\n // the result doesn't contain any suggestions.\n } else if (result.status == MLKSmartReplyResultStatusSuccess) {\n // Successfully suggested smart replies.\n // ...\n }\n}];\n```\n\nIf the operation succeeds, a `SmartReplySuggestionResult` object is passed to\nthe completion handler. This object contains a list of up to three suggested\nreplies, which you can present to your user: \n\nSwift \n\n```swift\nfor suggestion in result.suggestions {\n print(\"Suggested reply: \\(suggestion.text)\")\n}\n```\n\nObjective-C \n\n```objective-c\nfor (MLKSmartReplySuggestion *suggestion in result.suggestions) {\n NSLog(@\"Suggested reply: %@\", suggestion.text);\n}\n```\n\nNote that ML Kit might not return results if the model isn't confident in\nthe relevance of the suggested replies, the input conversation isn't in\nEnglish, or if the model detects sensitive subject matter."]]