快速入门：使用文本、文档和对话摘要

项目
09/02/2024

重要

对话摘要仅适用于：

REST API
Python
C#

参考文档 | 更多示例 | 包 (NuGet) | 库源代码

借助本快速入门，使用 .NET 客户端库创建文本摘要应用程序。在以下示例中，你将创建一个 C# 应用程序，该应用程序可以汇总文档或基于文本的客户服务对话。

提示

可以使用 Language Studio 体验文本摘要，而无需编写代码。

先决条件

Azure 订阅 - 创建试用订阅
Visual Studio IDE
拥有 Azure 订阅后，请在 Azure 门户中创建语言资源，以获取密钥和终结点。部署后，选择”转到资源”。
- 需要从创建的资源获取密钥和终结点，以便将应用程序连接到 API。稍后需要在本快速入门中将密钥和终结点粘贴到代码中。
- 可以使用免费定价层 (Free F0) 试用该服务，然后再升级到付费层进行生产。
若要使用“分析”功能，需要标准 (S) 定价层的“语言”资源。

设置

创建环境变量

应用程序必须经过身份验证才能发送 API 请求。对于生产，请使用安全的方式存储和访问凭据。在此示例中，你将凭据写入运行应用程序的本地计算机上的环境变量。

若要为语言资源密钥设置环境变量，请打开控制台窗口，并按照操作系统和开发环境的说明进行操作。

若要设置 LANGUAGE_KEY 环境变量，请将 your-key 替换为资源的其中一个密钥。
若要设置 LANGUAGE_ENDPOINT 环境变量，请将 your-endpoint 替换为资源的终结点。

重要

如果使用 API 密钥，请将其安全地存储在某个其他位置，例如 Azure Key Vault 中。请不要直接在代码中包含 API 密钥，并且切勿公开发布该密钥。

有关 Azure AI 服务安全性的详细信息，请参阅对 Azure AI 服务的请求进行身份验证。

setx LANGUAGE_KEY your-key

setx LANGUAGE_ENDPOINT your-endpoint

注意

如果只需要访问当前正在运行的控制台中的环境变量，则可以使用 set（而不是 setx）设置环境变量。

添加环境变量后，可能需要重启任何正在运行的、需要读取环境变量的程序（包括控制台窗口）。例如，如果使用 Visual Studio 作为编辑器，请在运行示例之前重启 Visual Studio。

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bashrc，使更改生效。

Bash

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bash_profile，使更改生效。

Xcode

对于 iOS 和 macOS 开发，可在 Xcode 中设置环境变量。例如，按照以下步骤在 Xcode 13.4.1 中设置环境变量。

选择“产品”>“方案”>“编辑方案”
在“运行(调试运行)”页上选择“参数”
在“环境变量”下，选择加号 (+) 符号来添加新的环境变量。
为“名称”输入 LANGUAGE_KEY，为“值”输入语言资源密钥。
为资源终结点执行这些步骤。将新的环境变量命名为 LANGUAGE_ENDPOINT。

有关更多配置选项，请参阅 Xcode 文档。

创建新的 .NET Core 应用程序

使用 Visual Studio IDE 创建新的 .NET Core 控制台应用。这会创建包含单个 C# 源文件的“Hello World”项目：program.cs。

右键单击解决方案资源管理器中的解决方案，然后选择“管理 NuGet 包”，以便安装客户端库。在打开的包管理器中选择“浏览”，搜索 Azure.AI.TextAnalytics。请务必选中“包括预发行版”。选择版本 5.3.0，然后选择“安装”。也可使用包管理器控制台。

代码示例

将以下代码复制到 program.cs 文件。然后运行代码。

重要

转到 Azure 门户。如果你在“先决条件”部分创建的语言资源部署成功，请单击“后续步骤”下的“转到资源”按钮。可以通过导航到资源的“密钥和终结点”页，在“资源管理”下找到你的密钥和终结点。

重要

完成后，请记住将密钥从代码中删除，并且永远不要公开发布该密钥。对于生产来说，请使用安全的方式存储和访问凭据，例如 Azure Key Vault。有关详细信息，请参阅 Azure AI 服务安全性一文。

using Azure;
using System;
using Azure.AI.TextAnalytics;
using System.Threading.Tasks;
using System.Collections.Generic;

namespace Example
{
    class Program
    {
        // This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
        static string languageKey = Environment.GetEnvironmentVariable("LANGUAGE_KEY");
        static string languageEndpoint = Environment.GetEnvironmentVariable("LANGUAGE_ENDPOINT");

        private static readonly AzureKeyCredential credentials = new AzureKeyCredential(languageKey);
        private static readonly Uri endpoint = new Uri(languageEndpoint);

        // Example method for summarizing text
        static async Task TextSummarizationExample(TextAnalyticsClient client)
        {
            string document = @"The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. 
                These sentences collectively convey the main idea of the document. This feature is provided as an API for developers. 
                They can use it to build intelligent solutions based on the relevant information extracted to support various use cases. 
                Extractive summarization supports several languages. It is based on pretrained multilingual transformer models, part of our quest for holistic representations. 
                It draws its strength from transfer learning across monolingual and harness the shared nature of languages to produce models of improved quality and efficiency." ;
        
            // Prepare analyze operation input. You can add multiple documents to this list and perform the same
            // operation to all of them.
            var batchInput = new List<string>
            {
                document
            };
        
            TextAnalyticsActions actions = new TextAnalyticsActions()
            {
                ExtractiveSummarizeActions = new List<ExtractiveSummarizeAction>() { new ExtractiveSummarizeAction() }
            };
        
            // Start analysis process.
            AnalyzeActionsOperation operation = await client.StartAnalyzeActionsAsync(batchInput, actions);
            await operation.WaitForCompletionAsync();
            // View operation status.
            Console.WriteLine($"AnalyzeActions operation has completed");
            Console.WriteLine();
        
            Console.WriteLine($"Created On   : {operation.CreatedOn}");
            Console.WriteLine($"Expires On   : {operation.ExpiresOn}");
            Console.WriteLine($"Id           : {operation.Id}");
            Console.WriteLine($"Status       : {operation.Status}");
        
            Console.WriteLine();
            // View operation results.
            await foreach (AnalyzeActionsResult documentsInPage in operation.Value)
            {
                IReadOnlyCollection<ExtractiveSummarizeActionResult> summaryResults = documentsInPage.ExtractiveSummarizeResults;
        
                foreach (ExtractiveSummarizeActionResult summaryActionResults in summaryResults)
                {
                    if (summaryActionResults.HasError)
                    {
                        Console.WriteLine($"  Error!");
                        Console.WriteLine($"  Action error code: {summaryActionResults.Error.ErrorCode}.");
                        Console.WriteLine($"  Message: {summaryActionResults.Error.Message}");
                        continue;
                    }
        
                    foreach (ExtractiveSummarizeResult documentResults in summaryActionResults.DocumentsResults)
                    {
                        if (documentResults.HasError)
                        {
                            Console.WriteLine($"  Error!");
                            Console.WriteLine($"  Document error code: {documentResults.Error.ErrorCode}.");
                            Console.WriteLine($"  Message: {documentResults.Error.Message}");
                            continue;
                        }
        
                        Console.WriteLine($"  Extracted the following {documentResults.Sentences.Count} sentence(s):");
                        Console.WriteLine();
        
                        foreach (ExtractiveSummarySentence sentence in documentResults.Sentences)
                        {
                            Console.WriteLine($"  Sentence: {sentence.Text}");
                            Console.WriteLine();
                        }
                    }
                }
            }
        }

        static async Task Main(string[] args)
        {
            var client = new TextAnalyticsClient(endpoint, credentials);
            await TextSummarizationExample(client);
        }
    }
}

输出

AnalyzeActions operation has completed

Created On   : 9/16/2021 8:04:27 PM +00:00
Expires On   : 9/17/2021 8:04:27 PM +00:00
Id           : 2e63fa58-fbaa-4be9-a700-080cff098f91
Status       : succeeded

Extracted the following 3 sentence(s):

Sentence: The extractive summarization feature in uses natural language processing techniques to locate key sentences in an unstructured text document.

Sentence: This feature is provided as an API for developers.

Sentence: They can use it to build intelligent solutions based on the relevant information extracted to support various use cases.

using System;
using Azure;
using Azure.Core;
using Azure.AI.Language.Conversations;
using System.Text.Json;
using System.Threading.Tasks;

namespace Example
{
    class Program
    {
        private static readonly AzureKeyCredential credentials = new AzureKeyCredential("replace-with-your-key-here");
        private static readonly Uri endpoint = new Uri("replace-with-your-endpoint-here");

        static async Task Main(string[] args)
        {
            ConversationAnalysisClient client = new ConversationAnalysisClient(endpoint, credentials);
            await Summarization(client);
        }

        static async Task Summarization(ConversationAnalysisClient client)
        {
            var data = new
            {
                analysisInput = new
                {
                    conversations = new[]
                    {
                        new
                        {
                            conversationItems = new[]
                            {
                                new
                                {
                                    text = "Hello, you’re chatting with Rene. How may I help you?",
                                    id = "1",
                                    role = "Agent",
                                    participantId = "Agent",
                                },
                                new
                                {
                                    text = "Hi, I tried to set up wifi connection for Smart Brew 300 coffee machine, but it didn’t work.",
                                    id = "2",
                                    role = "Customer",
                                    participantId = "Customer",
                                },
                                new
                                {
                                    text = "I’m sorry to hear that. Let’s see what we can do to fix this issue. Could you please try the following steps for me? First, could you push the wifi connection button, hold for 3 seconds, then let me know if the power light is slowly blinking on and off every second?",
                                    id = "3",
                                    role = "Agent",
                                    participantId = "Agent",
                                },
                                new
                                {
                                    text = "Yes, I pushed the wifi connection button, and now the power light is slowly blinking?",
                                    id = "4",
                                    role = "Customer",
                                    participantId = "Customer",
                                },
                                new
                                {
                                    text = "Great. Thank you! Now, please check in your Contoso Coffee app. Does it prompt to ask you to connect with the machine?",
                                    id = "5",
                                    role = "Agent",
                                    participantId = "Agent",
                                },
                                new
                                {
                                    text = "No. Nothing happened.",
                                    id = "6",
                                    role = "Customer",
                                    participantId = "Customer",
                                },
                                new
                                {
                                    text = "I’m very sorry to hear that. Let me see if there’s another way to fix the issue. Please hold on for a minute.",
                                    id = "7",
                                    role = "Agent",
                                    participantId = "Agent",
                                }
                            },
                            id = "1",
                            language = "en",
                            modality = "text",
                        },
                    }
                },
                tasks = new[]
                {
                    new
                    {
                        parameters = new
                        {
                            summaryAspects = new[]
                            {
                                "issue",
                                "resolution",
                            }
                        },
                        kind = "ConversationalSummarizationTask",
                        taskName = "1",
                    },
                },
            };

            Operation<BinaryData> analyzeConversationOperation = await client.AnalyzeConversationsAsync(WaitUntil.Started, RequestContent.Create(data));
            analyzeConversationOperation.WaitForCompletion();

            using JsonDocument result = JsonDocument.Parse(analyzeConversationOperation.Value.ToStream());
            JsonElement jobResults = result.RootElement;
            foreach (JsonElement task in jobResults.GetProperty("tasks").GetProperty("items").EnumerateArray())
            {
                JsonElement results = task.GetProperty("results");

                Console.WriteLine("Conversations:");
                foreach (JsonElement conversation in results.GetProperty("conversations").EnumerateArray())
                {
                    Console.WriteLine($"Conversation: #{conversation.GetProperty("id").GetString()}");
                    Console.WriteLine("Summaries:");
                    foreach (JsonElement summary in conversation.GetProperty("summaries").EnumerateArray())
                    {
                        Console.WriteLine($"Text: {summary.GetProperty("text").GetString()}");
                        Console.WriteLine($"Aspect: {summary.GetProperty("aspect").GetString()}");
                    }
                    Console.WriteLine();
                }
            }
        }
    }
}

输出

Conversations:
Conversation: #1
Summaries:
Text: Customer is unable to set up wifi connection for Smart Brew 300 coffee machine.
Aspect: issue
Text: The agent instructed the customer to press the wifi connection button, hold for 3 seconds, and check if the power light blinks. The agent also asked the customer to check if the Contoso Coffee app prompts to connect with the machine, but the issue persisted. The agent is looking for another way to fix the issue.
Aspect: resolution

using Azure;
using System;
using Azure.AI.TextAnalytics;
using System.Threading.Tasks;
using System.Collections.Generic;

namespace Example
{
    class Program
    {
        // This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
        static string languageKey = Environment.GetEnvironmentVariable("LANGUAGE_KEY");
        static string languageEndpoint = Environment.GetEnvironmentVariable("LANGUAGE_ENDPOINT");

        private static readonly AzureKeyCredential credentials = new AzureKeyCredential(languageKey);
        private static readonly Uri endpoint = new Uri(languageEndpoint);

        // Example method for summarizing text
        static async Task TextSummarizationExample(TextAnalyticsClient client)
        {
            string document = @"The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. 
                These sentences collectively convey the main idea of the document. This feature is provided as an API for developers. 
                They can use it to build intelligent solutions based on the relevant information extracted to support various use cases. 
                Extractive summarization supports several languages. It is based on pretrained multilingual transformer models, part of our quest for holistic representations. 
                It draws its strength from transfer learning across monolingual and harness the shared nature of languages to produce models of improved quality and efficiency." ;
        
            // Prepare analyze operation input. You can add multiple documents to this list and perform the same
            // operation to all of them.
            var batchInput = new List<string>
            {
                document
            };
        
            TextAnalyticsActions actions = new TextAnalyticsActions()
            {
                ExtractiveSummarizeActions = new List<ExtractiveSummarizeAction>() { new ExtractiveSummarizeAction() }
            };
        
            // Start analysis process.
            AnalyzeActionsOperation operation = await client.StartAnalyzeActionsAsync(batchInput, actions);
            await operation.WaitForCompletionAsync();
            // View operation status.
            Console.WriteLine($"AnalyzeActions operation has completed");
            Console.WriteLine();
        
            Console.WriteLine($"Created On   : {operation.CreatedOn}");
            Console.WriteLine($"Expires On   : {operation.ExpiresOn}");
            Console.WriteLine($"Id           : {operation.Id}");
            Console.WriteLine($"Status       : {operation.Status}");
        
            Console.WriteLine();
            // View operation results.
            await foreach (AnalyzeActionsResult documentsInPage in operation.Value)
            {
                IReadOnlyCollection<ExtractiveSummarizeActionResult> summaryResults = documentsInPage.ExtractiveSummarizeResults;
        
                foreach (ExtractiveSummarizeActionResult summaryActionResults in summaryResults)
                {
                    if (summaryActionResults.HasError)
                    {
                        Console.WriteLine($"  Error!");
                        Console.WriteLine($"  Action error code: {summaryActionResults.Error.ErrorCode}.");
                        Console.WriteLine($"  Message: {summaryActionResults.Error.Message}");
                        continue;
                    }
        
                    foreach (ExtractiveSummarizeResult documentResults in summaryActionResults.DocumentsResults)
                    {
                        if (documentResults.HasError)
                        {
                            Console.WriteLine($"  Error!");
                            Console.WriteLine($"  Document error code: {documentResults.Error.ErrorCode}.");
                            Console.WriteLine($"  Message: {documentResults.Error.Message}");
                            continue;
                        }
        
                        Console.WriteLine($"  Extracted the following {documentResults.Sentences.Count} sentence(s):");
                        Console.WriteLine();
        
                        foreach (ExtractiveSummarySentence sentence in documentResults.Sentences)
                        {
                            Console.WriteLine($"  Sentence: {sentence.Text}");
                            Console.WriteLine();
                        }
                    }
                }
            }
        }

        static async Task Main(string[] args)
        {
            var client = new TextAnalyticsClient(endpoint, credentials);
            await TextSummarizationExample(client);
        }
    }
}

输出

AnalyzeActions operation has completed

Created On   : 9/16/2021 8:04:27 PM +00:00
Expires On   : 9/17/2021 8:04:27 PM +00:00
Id           : 2e63fa58-fbaa-4be9-a700-080cff098f91
Status       : succeeded

Extracted the following 3 sentence(s):

Sentence: The extractive summarization feature in uses natural language processing techniques to locate key sentences in an unstructured text document.

Sentence: This feature is provided as an API for developers.

Sentence: They can use it to build intelligent solutions based on the relevant information extracted to support various use cases.

参考文档 | 更多示例 | 包 (Maven) | 库源代码

借助本快速入门，使用 Java 客户端库创建文本摘要应用程序。在以下示例中，你将创建一个可以汇总文档的 Java 应用程序。

提示

可以使用 Language Studio 体验文本摘要，而无需编写代码。

先决条件

Azure 订阅 - 创建试用订阅
Java 开发工具包 (JDK) 版本 8 或更高版本
拥有 Azure 订阅后，请在 Azure 门户中创建语言资源，以获取密钥和终结点。部署后，选择”转到资源”。
- 需要从创建的资源获取密钥和终结点，以便将应用程序连接到 API。稍后在快速入门中将密钥和终结点粘贴到下方的代码中。
- 可以使用免费定价层 (Free F0) 试用该服务，然后再升级到付费层进行生产。
若要使用“分析”功能，需要标准 (S) 定价层的“语言”资源。

设置

添加客户端库

在首选 IDE 或开发环境中创建 Maven 项目。然后在项目的 pom.xml 文件中，添加以下依赖项。可联机找到用于其他生成工具的实现语法。

<dependencies>
     <dependency>
        <groupId>com.azure</groupId>
        <artifactId>azure-ai-textanalytics</artifactId>
        <version>5.3.0</version>
    </dependency>
</dependencies>

创建环境变量

若要为语言资源密钥设置环境变量，请打开控制台窗口，并按照操作系统和开发环境的说明进行操作。

若要设置 LANGUAGE_KEY 环境变量，请将 your-key 替换为资源的其中一个密钥。
若要设置 LANGUAGE_ENDPOINT 环境变量，请将 your-endpoint 替换为资源的终结点。

重要

如果使用 API 密钥，请将其安全地存储在某个其他位置，例如 Azure Key Vault 中。请不要直接在代码中包含 API 密钥，并且切勿公开发布该密钥。

有关 Azure AI 服务安全性的详细信息，请参阅对 Azure AI 服务的请求进行身份验证。

setx LANGUAGE_KEY your-key

setx LANGUAGE_ENDPOINT your-endpoint

注意

如果只需要访问当前正在运行的控制台中的环境变量，则可以使用 set（而不是 setx）设置环境变量。

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bashrc，使更改生效。

Bash

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bash_profile，使更改生效。

Xcode

对于 iOS 和 macOS 开发，可在 Xcode 中设置环境变量。例如，按照以下步骤在 Xcode 13.4.1 中设置环境变量。

选择“产品”>“方案”>“编辑方案”
在“运行(调试运行)”页上选择“参数”
在“环境变量”下，选择加号 (+) 符号来添加新的环境变量。
为“名称”输入 LANGUAGE_KEY，为“值”输入语言资源密钥。
为资源终结点执行这些步骤。将新的环境变量命名为 LANGUAGE_ENDPOINT。

有关更多配置选项，请参阅 Xcode 文档。

代码示例

创建名为 Example.java 的 Java 文件。打开该文件，并复制以下代码。然后运行代码。

重要

import com.azure.core.credential.AzureKeyCredential;
import com.azure.ai.textanalytics.models.*;
import com.azure.ai.textanalytics.TextAnalyticsClientBuilder;
import com.azure.ai.textanalytics.TextAnalyticsClient;
import java.util.ArrayList;
import java.util.List;
import com.azure.core.util.polling.SyncPoller;
import com.azure.ai.textanalytics.util.*;

public class Example {

    // This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
    private static String languageKey = System.getenv("LANGUAGE_KEY");
    private static String languageEndpoint = System.getenv("LANGUAGE_ENDPOINT");

    public static void main(String[] args) {
        TextAnalyticsClient client = authenticateClient(languageKey, languageEndpoint);
        summarizationExample(client);
    }
    // Method to authenticate the client object with your key and endpoint
    static TextAnalyticsClient authenticateClient(String key, String endpoint) {
        return new TextAnalyticsClientBuilder()
                .credential(new AzureKeyCredential(key))
                .endpoint(endpoint)
                .buildClient();
    }
    // Example method for summarizing text
    static void summarizationExample(TextAnalyticsClient client) {
        List<String> documents = new ArrayList<>();
        documents.add(
                "The extractive summarization feature uses natural language processing techniques "
                + "to locate key sentences in an unstructured text document. "
                + "These sentences collectively convey the main idea of the document. This feature is provided as an API for developers. "
                + "They can use it to build intelligent solutions based on the relevant information extracted to support various use cases. "
                + "Extractive summarization supports several languages. "
                + "It is based on pretrained multilingual transformer models, part of our quest for holistic representations. "
                + "It draws its strength from transfer learning across monolingual and harness the shared nature of languages "
                + "to produce models of improved quality and efficiency.");
    
        SyncPoller<AnalyzeActionsOperationDetail, AnalyzeActionsResultPagedIterable> syncPoller =
                client.beginAnalyzeActions(documents,
                        new TextAnalyticsActions().setDisplayName("{tasks_display_name}")
                                .setExtractSummaryActions(
                                        new ExtractSummaryAction()),
                        "en",
                        new AnalyzeActionsOptions());
    
        syncPoller.waitForCompletion();
    
        syncPoller.getFinalResult().forEach(actionsResult -> {
            System.out.println("Extractive Summarization action results:");
            for (ExtractSummaryActionResult actionResult : actionsResult.getExtractSummaryResults()) {
                if (!actionResult.isError()) {
                    for (ExtractSummaryResult documentResult : actionResult.getDocumentsResults()) {
                        if (!documentResult.isError()) {
                            System.out.println("\tExtracted summary sentences:");
                            for (SummarySentence summarySentence : documentResult.getSentences()) {
                                System.out.printf(
                                        "\t\t Sentence text: %s, length: %d, offset: %d, rank score: %f.%n",
                                        summarySentence.getText(), summarySentence.getLength(),
                                        summarySentence.getOffset(), summarySentence.getRankScore());
                            }
                        } else {
                            System.out.printf("\tCannot extract summary sentences. Error: %s%n",
                                    documentResult.getError().getMessage());
                        }
                    }
                } else {
                    System.out.printf("\tCannot execute Extractive Summarization action. Error: %s%n",
                            actionResult.getError().getMessage());
                }
            }
        });
    }
}

输出

Extractive Summarization action results:
	Extracted summary sentences:
		 Sentence text: The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document., length: 138, offset: 0, rank score: 1.000000.
		 Sentence text: This feature is provided as an API for developers., length: 50, offset: 206, rank score: 0.510000.
		 Sentence text: Extractive summarization supports several languages., length: 52, offset: 378, rank score: 0.410000.

参考文档 | 其他示例 | 包 (npm) | 库源代码

借助本快速入门，使用 Node.js 客户端库创建文本摘要应用程序。在以下示例中，你将创建一个可以汇总文档的 JavaScript 应用程序。

提示

可以使用 Language Studio 体验文本摘要，而无需编写代码。

先决条件

Azure 订阅 - 创建试用订阅
Node.js v16 LTS
拥有 Azure 订阅后，请在 Azure 门户中创建语言资源，以获取密钥和终结点。部署后，选择”转到资源”。
- 需要从创建的资源获取密钥和终结点，以便将应用程序连接到 API。你稍后会在快速入门中将密钥和终结点粘贴到下方的代码中。
- 可以使用免费定价层 (Free F0) 试用该服务，然后再升级到付费层进行生产。
若要使用“分析”功能，需要标准 (S) 定价层的“语言”资源。

设置

创建环境变量

若要为语言资源密钥设置环境变量，请打开控制台窗口，并按照操作系统和开发环境的说明进行操作。

若要设置 LANGUAGE_KEY 环境变量，请将 your-key 替换为资源的其中一个密钥。
若要设置 LANGUAGE_ENDPOINT 环境变量，请将 your-endpoint 替换为资源的终结点。

重要

如果使用 API 密钥，请将其安全地存储在某个其他位置，例如 Azure Key Vault 中。请不要直接在代码中包含 API 密钥，并且切勿公开发布该密钥。

有关 Azure AI 服务安全性的详细信息，请参阅对 Azure AI 服务的请求进行身份验证。

setx LANGUAGE_KEY your-key

setx LANGUAGE_ENDPOINT your-endpoint

注意

如果只需要访问当前正在运行的控制台中的环境变量，则可以使用 set（而不是 setx）设置环境变量。

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bashrc，使更改生效。

Bash

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bash_profile，使更改生效。

Xcode

对于 iOS 和 macOS 开发，可在 Xcode 中设置环境变量。例如，按照以下步骤在 Xcode 13.4.1 中设置环境变量。

选择“产品”>“方案”>“编辑方案”
在“运行(调试运行)”页上选择“参数”
在“环境变量”下，选择加号 (+) 符号来添加新的环境变量。
为“名称”输入 LANGUAGE_KEY，为“值”输入语言资源密钥。
为资源终结点执行这些步骤。将新的环境变量命名为 LANGUAGE_ENDPOINT。

有关更多配置选项，请参阅 Xcode 文档。

创建新的 Node.js 应用程序

在控制台窗口（例如 cmd、PowerShell 或 Bash）中，为应用创建一个新目录并导航到该目录。

mkdir myapp 

cd myapp

运行 npm init 命令以使用 package.json 文件创建一个 node 应用程序。

npm init

安装客户端库

安装 npm 包：

npm install --save @azure/ai-language-text@1.1.0

代码示例

打开该文件，并复制以下代码。然后运行代码。

重要

/**
 * This sample program extracts a summary of two sentences at max from an article.
 * For more information, see the feature documentation: {@link https://learn.microsoft.com/azure/ai-services/language-service/summarization/overview}
 *
 * @summary extracts a summary from an article
 */

const { AzureKeyCredential, TextAnalysisClient } = require("@azure/ai-language-text");

// Load the .env file if it exists
require("dotenv").config();

// This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
const endpoint = process.env.LANGUAGE_ENDPOINT;
const apiKey = process.env.LANGUAGE_KEY;

const documents = [
  `
           Windows 365 was in the works before COVID-19 sent companies around the world on a scramble to secure solutions to support employees suddenly forced to work from home, but “what really put the firecracker behind it was the pandemic, it accelerated everything,” McKelvey said. She explained that customers were asking, “’How do we create an experience for people that makes them still feel connected to the company without the physical presence of being there?”
           In this new world of Windows 365, remote workers flip the lid on their laptop, bootup the family workstation or clip a keyboard onto a tablet, launch a native app or modern web browser and login to their Windows 365 account. From there, their Cloud PC appears with their background, apps, settings and content just as they left it when they last were last there - in the office, at home or a coffee shop.
           “And then, when you’re done, you’re done. You won’t have any issues around security because you’re not saving anything on your device,” McKelvey said, noting that all the data is stored in the cloud.
           The ability to login to a Cloud PC from anywhere on any device is part of Microsoft’s larger strategy around tailoring products such as Microsoft Teams and Microsoft 365 for the post-pandemic hybrid workforce of the future, she added. It enables employees accustomed to working from home to continue working from home; it enables companies to hire interns from halfway around the world; it allows startups to scale without requiring IT expertise.
           “I think this will be interesting for those organizations who, for whatever reason, have shied away from virtualization. This is giving them an opportunity to try it in a way that their regular, everyday endpoint admin could manage,” McKelvey said.
           The simplicity of Windows 365 won over Dean Wells, the corporate chief information officer for the Government of Nunavut. His team previously attempted to deploy a traditional virtual desktop infrastructure and found it inefficient and unsustainable given the limitations of low-bandwidth satellite internet and the constant need for IT staff to manage the network and infrastructure.
           We didn’t run it for very long,” he said. “It didn’t turn out the way we had hoped. So, we actually had terminated the project and rolled back out to just regular PCs.”
           He re-evaluated this decision after the Government of Nunavut was hit by a ransomware attack in November 2019 that took down everything from the phone system to the government’s servers. Microsoft helped rebuild the system, moving the government to Teams, SharePoint, OneDrive and Microsoft 365. Manchester’s team recruited the Government of Nunavut to pilot Windows 365. Wells was intrigued, especially by the ability to manage the elastic workforce securely and seamlessly.
           “The impact that I believe we are finding, and the impact that we’re going to find going forward, is being able to access specialists from outside the territory and organizations outside the territory to come in and help us with our projects, being able to get people on staff with us to help us deliver the day-to-day expertise that we need to run the government,” he said.
           “Being able to improve healthcare, being able to improve education, economic development is going to improve the quality of life in the communities.”`,
];

async function main() {
  console.log("== Extractive Summarization Sample ==");

  const client = new TextAnalysisClient(endpoint, new AzureKeyCredential(apiKey));
  const actions = [
    {
      kind: "ExtractiveSummarization",
      maxSentenceCount: 2,
    },
  ];
  const poller = await client.beginAnalyzeBatch(actions, documents, "en");

  poller.onProgress(() => {
    console.log(
      `Last time the operation was updated was on: ${poller.getOperationState().modifiedOn}`
    );
  });
  console.log(`The operation was created on ${poller.getOperationState().createdOn}`);
  console.log(`The operation results will expire on ${poller.getOperationState().expiresOn}`);

  const results = await poller.pollUntilDone();

  for await (const actionResult of results) {
    if (actionResult.kind !== "ExtractiveSummarization") {
      throw new Error(`Expected extractive summarization results but got: ${actionResult.kind}`);
    }
    if (actionResult.error) {
      const { code, message } = actionResult.error;
      throw new Error(`Unexpected error (${code}): ${message}`);
    }
    for (const result of actionResult.results) {
      console.log(`- Document ${result.id}`);
      if (result.error) {
        const { code, message } = result.error;
        throw new Error(`Unexpected error (${code}): ${message}`);
      }
      console.log("Summary:");
      console.log(result.sentences.map((sentence) => sentence.text).join("\n"));
    }
  }
}

main().catch((err) => {
  console.error("The sample encountered an error:", err);
});

module.exports = { main };

参考文档 | 更多示例 | 包 (PyPi) | 库源代码

借助本快速入门，使用 Python 客户端库创建文本摘要应用程序。在以下示例中，你将创建一个 Python 应用程序，该应用程序可以汇总文档或基于文本的客户服务对话。

提示

可以使用 Language Studio 体验文本摘要，而无需编写代码。

先决条件

Azure 订阅 - 创建试用订阅
Python 3.x
拥有 Azure 订阅后，请在 Azure 门户中创建语言资源，以获取密钥和终结点。部署后，选择”转到资源”。
- 需要从创建的资源获取密钥和终结点，以便将应用程序连接到 API。稍后在快速入门中将密钥和终结点粘贴到下方的代码中。
- 可以使用免费定价层 (Free F0) 试用该服务，然后再升级到付费层进行生产。
若要使用“分析”功能，需要标准 (S) 定价层的“语言”资源。

设置

创建环境变量

若要为语言资源密钥设置环境变量，请打开控制台窗口，并按照操作系统和开发环境的说明进行操作。

若要设置 LANGUAGE_KEY 环境变量，请将 your-key 替换为资源的其中一个密钥。
若要设置 LANGUAGE_ENDPOINT 环境变量，请将 your-endpoint 替换为资源的终结点。

重要

如果使用 API 密钥，请将其安全地存储在某个其他位置，例如 Azure Key Vault 中。请不要直接在代码中包含 API 密钥，并且切勿公开发布该密钥。

有关 Azure AI 服务安全性的详细信息，请参阅对 Azure AI 服务的请求进行身份验证。

setx LANGUAGE_KEY your-key

setx LANGUAGE_ENDPOINT your-endpoint

注意

如果只需要访问当前正在运行的控制台中的环境变量，则可以使用 set（而不是 setx）设置环境变量。

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bashrc，使更改生效。

Bash

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bash_profile，使更改生效。

Xcode

对于 iOS 和 macOS 开发，可在 Xcode 中设置环境变量。例如，按照以下步骤在 Xcode 13.4.1 中设置环境变量。

选择“产品”>“方案”>“编辑方案”
在“运行(调试运行)”页上选择“参数”
在“环境变量”下，选择加号 (+) 符号来添加新的环境变量。
为“名称”输入 LANGUAGE_KEY，为“值”输入语言资源密钥。
为资源终结点执行这些步骤。将新的环境变量命名为 LANGUAGE_ENDPOINT。

有关更多配置选项，请参阅 Xcode 文档。

安装客户端库

在安装 Python 后，可以通过以下命令安装客户端库：

pip install azure-ai-textanalytics==5.3.0

pip install azure-ai-language-conversations==1.1.0

pip install azure-ai-textanalytics==5.3.0

代码示例

创建新的 Python 文件，并复制以下代码。然后运行代码。

重要

# This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
key = os.environ.get('LANGUAGE_KEY')
endpoint = os.environ.get('LANGUAGE_ENDPOINT')

from azure.ai.textanalytics import TextAnalyticsClient
from azure.core.credentials import AzureKeyCredential

# Authenticate the client using your key and endpoint 
def authenticate_client():
    ta_credential = AzureKeyCredential(key)
    text_analytics_client = TextAnalyticsClient(
            endpoint=endpoint, 
            credential=ta_credential)
    return text_analytics_client

client = authenticate_client()

# Example method for summarizing text
def sample_extractive_summarization(client):
    from azure.core.credentials import AzureKeyCredential
    from azure.ai.textanalytics import (
        TextAnalyticsClient,
        ExtractiveSummaryAction
    ) 

    document = [
        "The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. "
        "These sentences collectively convey the main idea of the document. This feature is provided as an API for developers. " 
        "They can use it to build intelligent solutions based on the relevant information extracted to support various use cases. "
        "Extractive summarization supports several languages. It is based on pretrained multilingual transformer models, part of our quest for holistic representations. "
        "It draws its strength from transfer learning across monolingual and harness the shared nature of languages to produce models of improved quality and efficiency. "
    ]

    poller = client.begin_analyze_actions(
        document,
        actions=[
            ExtractiveSummaryAction(max_sentence_count=4)
        ],
    )

    document_results = poller.result()
    for result in document_results:
        extract_summary_result = result[0]  # first document, first result
        if extract_summary_result.is_error:
            print("...Is an error with code '{}' and message '{}'".format(
                extract_summary_result.code, extract_summary_result.message
            ))
        else:
            print("Summary extracted: \n{}".format(
                " ".join([sentence.text for sentence in extract_summary_result.sentences]))
            )

sample_extractive_summarization(client)

输出

Summary extracted: 
The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. This feature is provided as an API for developers. They can use it to build intelligent solutions based on the relevant information extracted to support various use cases.


key = "paste-your-key-here"
endpoint = "paste-your-endpoint-here"

import os
from azure.core.credentials import AzureKeyCredential
from azure.ai.language.conversations import ConversationAnalysisClient

client = ConversationAnalysisClient(endpoint, AzureKeyCredential(key))
with client:
    poller = client.begin_conversation_analysis(
        task={
            "displayName": "Analyze conversations from xxx",
            "analysisInput": {
                "conversations": [
                    {
                        "conversationItems": [
                            {
                                "text": "Hello, you’re chatting with Rene. How may I help you?",
                                "id": "1",
                                "role": "Agent",
                                "participantId": "Agent_1",
                            },
                            {
                                "text": "Hi, I tried to set up wifi connection for Smart Brew 300 coffee machine, but it didn’t work.",
                                "id": "2",
                                "role": "Customer",
                                "participantId": "Customer_1",
                            },
                            {
                                "text": "I’m sorry to hear that. Let’s see what we can do to fix this issue. Could you please try the following steps for me? First, could you push the wifi connection button, hold for 3 seconds, then let me know if the power light is slowly blinking on and off every second?",
                                "id": "3",
                                "role": "Agent",
                                "participantId": "Agent_1",
                            },
                            {
                                "text": "Yes, I pushed the wifi connection button, and now the power light is slowly blinking.",
                                "id": "4",
                                "role": "Customer",
                                "participantId": "Customer_1",
                            },
                            {
                                "text": "Great. Thank you! Now, please check in your Contoso Coffee app. Does it prompt to ask you to connect with the machine?",
                                "id": "5",
                                "role": "Agent",
                                "participantId": "Agent_1",
                            },
                            {
                                "text": "No. Nothing happened.",
                                "id": "6",
                                "role": "Customer",
                                "participantId": "Customer_1",
                            },
                            {
                                "text": "I’m very sorry to hear that. Let me see if there’s another way to fix the issue. Please hold on for a minute.",
                                "id": "7",
                                "role": "Agent",
                                "participantId": "Agent_1",
                            },
                        ],
                        "modality": "text",
                        "id": "conversation1",
                        "language": "en",
                    },
                ]
            },
            "tasks": [
                {
                    "taskName": "Issue task",
                    "kind": "ConversationalSummarizationTask",
                    "parameters": {"summaryAspects": ["issue"]},
                },
                {
                    "taskName": "Resolution task",
                    "kind": "ConversationalSummarizationTask",
                    "parameters": {"summaryAspects": ["resolution"]},
                },
            ],
        }
    )

    # view result
    result = poller.result()
    task_results = result["tasks"]["items"]
    for task in task_results:
        print(f"\n{task['taskName']} status: {task['status']}")
        task_result = task["results"]
        if task_result["errors"]:
            print("... errors occurred ...")
            for error in task_result["errors"]:
                print(error)
        else:
            conversation_result = task_result["conversations"][0]
            if conversation_result["warnings"]:
                print("... view warnings ...")
                for warning in conversation_result["warnings"]:
                    print(warning)
            else:
                summaries = conversation_result["summaries"]
                for summary in summaries:
                    print(f"{summary['aspect']}: {summary['text']}")

输出

Issue task status: succeeded
issue: Customer tried to set up wifi connection for Smart Brew 300 coffee machine but it didn't work. No error message.

Resolution task status: succeeded
resolution: Asked customer to check if the Contoso Coffee app prompts to connect with the machine. Customer ended the chat.

# This example requires environment variables named "LANGUAGE_KEY" and "LANGUAGE_ENDPOINT"
key = os.environ.get('LANGUAGE_KEY')
endpoint = os.environ.get('LANGUAGE_ENDPOINT')

from azure.ai.textanalytics import TextAnalyticsClient
from azure.core.credentials import AzureKeyCredential

# Authenticate the client using your key and endpoint 
def authenticate_client():
    ta_credential = AzureKeyCredential(key)
    text_analytics_client = TextAnalyticsClient(
            endpoint=endpoint, 
            credential=ta_credential)
    return text_analytics_client

client = authenticate_client()

# Example method for summarizing text
def sample_extractive_summarization(client):
    from azure.core.credentials import AzureKeyCredential
    from azure.ai.textanalytics import (
        TextAnalyticsClient,
        ExtractiveSummaryAction
    ) 

    document = [
        "The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. "
        "These sentences collectively convey the main idea of the document. This feature is provided as an API for developers. " 
        "They can use it to build intelligent solutions based on the relevant information extracted to support various use cases. "
        "Extractive summarization supports several languages. It is based on pretrained multilingual transformer models, part of our quest for holistic representations. "
        "It draws its strength from transfer learning across monolingual and harness the shared nature of languages to produce models of improved quality and efficiency. "
    ]

    poller = client.begin_analyze_actions(
        document,
        actions=[
            ExtractiveSummaryAction(max_sentence_count=4)
        ],
    )

    document_results = poller.result()
    for result in document_results:
        extract_summary_result = result[0]  # first document, first result
        if extract_summary_result.is_error:
            print("...Is an error with code '{}' and message '{}'".format(
                extract_summary_result.code, extract_summary_result.message
            ))
        else:
            print("Summary extracted: \n{}".format(
                " ".join([sentence.text for sentence in extract_summary_result.sentences]))
            )

sample_extractive_summarization(client)

输出

Summary extracted: 
The extractive summarization feature uses natural language processing techniques to locate key sentences in an unstructured text document. This feature is provided as an API for developers. They can use it to build intelligent solutions based on the relevant information extracted to support various use cases.

借助本快速入门，使用 REST API 发送文本摘要请求。在以下示例中，你将使用 cURL 来汇总文档或基于文本的客户服务对话。

提示

可以使用 Language Studio 体验文本摘要，而无需编写代码。

先决条件

最新版本的 cURL。
拥有 Azure 订阅后，请在 Azure 门户中创建语言资源，以获取密钥和终结点。部署后，选择”转到资源”。
- 需要从创建的资源获取密钥和终结点，以便将应用程序连接到 API。你稍后会在快速入门中将密钥和终结点粘贴到下方的代码中。
- 可以使用免费定价层 (Free F0) 试用该服务，然后再升级到付费层进行生产。

设置

创建环境变量

若要为语言资源密钥设置环境变量，请打开控制台窗口，并按照操作系统和开发环境的说明进行操作。

若要设置 LANGUAGE_KEY 环境变量，请将 your-key 替换为资源的其中一个密钥。
若要设置 LANGUAGE_ENDPOINT 环境变量，请将 your-endpoint 替换为资源的终结点。

重要

如果使用 API 密钥，请将其安全地存储在某个其他位置，例如 Azure Key Vault 中。请不要直接在代码中包含 API 密钥，并且切勿公开发布该密钥。

有关 Azure AI 服务安全性的详细信息，请参阅对 Azure AI 服务的请求进行身份验证。

setx LANGUAGE_KEY your-key

setx LANGUAGE_ENDPOINT your-endpoint

注意

如果只需要访问当前正在运行的控制台中的环境变量，则可以使用 set（而不是 setx）设置环境变量。

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bashrc，使更改生效。

Bash

export LANGUAGE_KEY=your-key

export LANGUAGE_ENDPOINT=your-endpoint

添加环境变量后，请从控制台窗口运行 source ~/.bash_profile，使更改生效。

Xcode

对于 iOS 和 macOS 开发，可在 Xcode 中设置环境变量。例如，按照以下步骤在 Xcode 13.4.1 中设置环境变量。

选择“产品”>“方案”>“编辑方案”
在“运行(调试运行)”页上选择“参数”
在“环境变量”下，选择加号 (+) 符号来添加新的环境变量。
为“名称”输入 LANGUAGE_KEY，为“值”输入语言资源密钥。
为资源终结点执行这些步骤。将新的环境变量命名为 LANGUAGE_ENDPOINT。

有关更多配置选项，请参阅 Xcode 文档。

示例请求

注意

以下 BASH 示例使用 \ 行继续符。如果你的控制台或终端使用不同的行继续符，请使用该字符。
可以在 GitHub 上找到特定于语言的示例。若要调用 API，需要以下信息：

选择要执行的摘要类型，然后选择以下选项卡之一来查看示例 API 调用：

功能	说明
文本摘要	使用抽取式文本摘要生成文档中重要或相关信息的摘要。
对话摘要	使用抽象式文本摘要生成客户服务代理与客户之间的口述文本中的问题和解决方法。

parameter	说明
`-X POST <endpoint>`	指定用于访问 API 的终结点。
`-H Content-Type: application/json`	用于发送 JSON 数据的内容类型。
`-H "Ocp-Apim-Subscription-Key:<key>`	指定用于访问 API 的密钥。
`-d <documents>`	包含要发送的文档的 JSON。

以下 cURL 命令从 BASH shell 中执行。使用你自己的 JSON 值编辑这些命令。

文本摘要

文本抽取式摘要示例

以下示例将帮助你开始使用文本抽取式摘要：

将下面的命令复制到文本编辑器中。 BASH 示例使用 \ 续行符。如果你的控制台或终端使用不同的续行符，请改用该续行符。

curl -i -X POST $LANGUAGE_ENDPOINT/language/analyze-text/jobs?api-version=2023-04-01 \
-H "Content-Type: application/json" \
-H "Ocp-Apim-Subscription-Key: $LANGUAGE_KEY" \
-d \
' 
{
  "displayName": "Text ext Summarization Task Example",
  "analysisInput": {
    "documents": [
      {
        "id": "1",
        "language": "en",
        "text": "At Microsoft, we have been on a quest to advance AI beyond existing techniques, by taking a more holistic, human-centric approach to learning and understanding. As Chief Technology Officer of Azure AI services, I have been working with a team of amazing scientists and engineers to turn this quest into a reality. In my role, I enjoy a unique perspective in viewing the relationship among three attributes of human cognition: monolingual text (X), audio or visual sensory signals, (Y) and multilingual (Z). At the intersection of all three, there’s magic—what we call XYZ-code as illustrated in Figure 1—a joint representation to create more powerful AI that can speak, hear, see, and understand humans better. We believe XYZ-code will enable us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages. The goal is to have pre-trained models that can jointly learn representations to support a broad range of downstream AI tasks, much in the way humans do today. Over the past five years, we have achieved human performance on benchmarks in conversational speech recognition, machine translation, conversational question answering, machine reading comprehension, and image captioning. These five breakthroughs provided us with strong signals toward our more ambitious aspiration to produce a leap in AI capabilities, achieving multi-sensory and multilingual learning that is closer in line with how humans learn and understand. I believe the joint XYZ-code is a foundational component of this aspiration, if grounded with external knowledge sources in the downstream AI tasks."
      }
    ]
  },
  "tasks": [
    {
      "kind": "ExtractiveSummarization",
      "taskName": "Text Extractive Summarization Task 1",
      "parameters": {
        "sentenceCount": 6
      }
    }
  ]
}
'

打开命令提示符窗口（例如 BASH）。
将文本编辑器中的命令粘贴到命令提示符窗口，然后运行该命令。
从响应头获取 operation-location。值将如下 URL 所示：

https://<your-language-resource-endpoint>/language/analyze-text/jobs/12345678-1234-1234-1234-12345678?api-version=2023-04-01

要获取请求的结果，请使用以下 cURL 命令。请务必将 <my-job-id> 替换为从之前的 operation-location 响应头中收到的数值 ID 值：

curl -X GET $LANGUAGE_ENDPOINT/language/analyze-text/jobs/<my-job-id>?api-version=2023-04-01 \
-H "Content-Type: application/json" \
-H "Ocp-Apim-Subscription-Key: $LANGUAGE_KEY"

文本抽取式摘要示例 JSON 响应

{
    "jobId": "56e43bcf-70d8-44d2-a7a7-131f3dff069f",
    "lastUpdateDateTime": "2022-09-28T19:33:43Z",
    "createdDateTime": "2022-09-28T19:33:42Z",
    "expirationDateTime": "2022-09-29T19:33:42Z",
    "status": "succeeded",
    "errors": [],
    "displayName": "Text ext Summarization Task Example",
    "tasks": {
        "completed": 1,
        "failed": 0,
        "inProgress": 0,
        "total": 1,
        "items": [
            {
                "kind": "ExtractiveSummarizationLROResults",
                "taskName": "Text Extractive Summarization Task 1",
                "lastUpdateDateTime": "2022-09-28T19:33:43.6712507Z",
                "status": "succeeded",
                "results": {
                    "documents": [
                        {
                            "id": "1",
                            "sentences": [
                                {
                                    "text": "At Microsoft, we have been on a quest to advance AI beyond existing techniques, by taking a more holistic, human-centric approach to learning and understanding.",
                                    "rankScore": 0.69,
                                    "offset": 0,
                                    "length": 160
                                },
                                {
                                    "text": "In my role, I enjoy a unique perspective in viewing the relationship among three attributes of human cognition: monolingual text (X), audio or visual sensory signals, (Y) and multilingual (Z).",
                                    "rankScore": 0.66,
                                    "offset": 324,
                                    "length": 192
                                },
                                {
                                    "text": "At the intersection of all three, there’s magic—what we call XYZ-code as illustrated in Figure 1—a joint representation to create more powerful AI that can speak, hear, see, and understand humans better.",
                                    "rankScore": 0.63,
                                    "offset": 517,
                                    "length": 203
                                },
                                {
                                    "text": "We believe XYZ-code will enable us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages.",
                                    "rankScore": 1.0,
                                    "offset": 721,
                                    "length": 134
                                },
                                {
                                    "text": "The goal is to have pre-trained models that can jointly learn representations to support a broad range of downstream AI tasks, much in the way humans do today.",
                                    "rankScore": 0.74,
                                    "offset": 856,
                                    "length": 159
                                },
                                {
                                    "text": "I believe the joint XYZ-code is a foundational component of this aspiration, if grounded with external knowledge sources in the downstream AI tasks.",
                                    "rankScore": 0.49,
                                    "offset": 1481,
                                    "length": 148
                                }
                            ],
                            "warnings": []
                        }
                    ],
                    "errors": [],
                    "modelVersion": "latest"
                }
            }
        ]
    }
}

对话问题和解决方法摘要

借助以下示例，你可以开始了解对话问题和解决方法摘要：

将下面的命令复制到文本编辑器中。 BASH 示例使用 \ 续行符。如果你的控制台或终端使用不同的续行符，请改用该续行符。

curl -i -X POST $LANGUAGE_ENDPOINT/language/analyze-conversations/jobs?api-version=2023-04-01 \
-H "Content-Type: application/json" \
-H "Ocp-Apim-Subscription-Key: $LANGUAGE_KEY" \
-d \
' 
{
  "displayName": "Conversation Task Example",
  "analysisInput": {
    "conversations": [
      {
        "conversationItems": [
          {
            "text": "Hello, you’re chatting with Rene. How may I help you?",
            "id": "1",
            "role": "Agent",
            "participantId": "Agent_1"
          },
          {
            "text": "Hi, I tried to set up wifi connection for Smart Brew 300 espresso machine, but it didn’t work.",
            "id": "2",
            "role": "Customer",
            "participantId": "Customer_1"
          },
          {
            "text": "I’m sorry to hear that. Let’s see what we can do to fix this issue. Could you please try the following steps for me? First, could you push the wifi connection button, hold for 3 seconds, then let me know if the power light is slowly blinking on and off every second?",
            "id": "3",
            "role": "Agent",
            "participantId": "Agent_1"
          },
          {
            "text": "Yes, I pushed the wifi connection button, and now the power light is slowly blinking.",
            "id": "4",
            "role": "Customer",
            "participantId": "Customer_1"
          },
          {
            "text": "Great. Thank you! Now, please check in your Contoso Coffee app. Does it prompt to ask you to connect with the machine? ",
            "id": "5",
            "role": "Agent",
            "participantId": "Agent_1"
          },
          {
            "text": "No. Nothing happened.",
            "id": "6",
            "role": "Customer",
            "participantId": "Customer_1"
          },
          {
            "text": "I’m very sorry to hear that. Let me see if there’s another way to fix the issue. Please hold on for a minute.",
            "id": "7",
            "role": "Agent",
            "participantId": "Agent_1"
          }
        ],
        "modality": "text",
        "id": "conversation1",
        "language": "en"
      }
    ]
  },
  "tasks": [
    {
      "taskName": "Conversation Task 1",
      "kind": "ConversationalSummarizationTask",
      "parameters": {
        "summaryAspects": ["issue"]
      }
    },
    {
      "taskName": "Conversation Task 2",
      "kind": "ConversationalSummarizationTask",
      "parameters": {
        "summaryAspects": ["resolution"],
        "sentenceCount": 1
      }
    }
  ]
}
'

只有 resolution 纵横比支持 sentenceCount。如果未指定 sentenceCount 参数，则模型将确定摘要的长度。请注意，sentenceCount 只是输出摘要的句子计数的近似值，范围为 1 到 7。

打开命令提示符窗口（例如 BASH）。
将文本编辑器中的命令粘贴到命令提示符窗口，然后运行该命令。
从响应头获取 operation-location。值将如下 URL 所示：

https://<your-language-resource-endpoint>/language/analyze-conversations/jobs/12345678-1234-1234-1234-12345678?api-version=2023-04-01

要获取请求的结果，请使用以下 cURL 命令。请务必将 <my-job-id> 替换为从之前的 operation-location 响应头中收到的数值 ID 值：

curl -X GET $LANGUAGE_ENDPOINT/language/analyze-conversations/jobs/<my-job-id>?api-version=2023-04-01 \
-H "Content-Type: application/json" \
-H "Ocp-Apim-Subscription-Key: $LANGUAGE_KEY"

对话问题和解决方法摘要示例 JSON 响应

{
  "jobId": "02ec5134-78bf-45da-8f63-d7410291ec40",
  "lastUpdatedDateTime": "2022-09-29T17:43:02Z",
  "createdDateTime": "2022-09-29T17:43:01Z",
  "expirationDateTime": "2022-09-30T17:43:01Z",
  "status": "succeeded",
  "errors": [],
  "displayName": "Conversation Task Example",
  "tasks": {
    "completed": 2,
    "failed": 0,
    "inProgress": 0,
    "total": 2,
    "items": [
      {
        "kind": "conversationalSummarizationResults",
        "taskName": "Conversation Task 1",
        "lastUpdateDateTime": "2022-09-29T17:43:02.3584219Z",
        "status": "succeeded",
        "results": {
          "conversations": [
            {
              "summaries": [
                {
                  "aspect": "issue",
                  "text": "Customer wants to connect their Smart Brew 300 to their Wi-Fi. | The Wi-Fi connection didn't work."
                }
              ],
              "id": "conversation1",
              "warnings": []
            }
          ],
          "errors": [],
          "modelVersion": "latest"
        }
      },
      {
        "kind": "conversationalSummarizationResults",
        "taskName": "Conversation Task 2",
        "lastUpdateDateTime": "2022-09-29T17:43:02.2099663Z",
        "status": "succeeded",
        "results": {
          "conversations": [
            {
              "summaries": [
                {
                  "aspect": "resolution",
                  "text": "Asked customer to check if the power light is blinking on and off every second."
                }
              ],
              "id": "conversation1",
              "warnings": []
            }
          ],
          "errors": [],
          "modelVersion": "latest"
        }
      }
    ]
  }
}

摘要示例文档

对于此项目，你需要将源文档上传到源容器。可下载本快速入门的 Microsoft Word 示例文档或 Adobe PDF。源语言为英语。

生成 POST 请求

使用首选编辑器或 IDE，为应用创建一个名为 native-document 的新目录。
在 native-document 目录中创建名为 document-summarization.json 的新 json 文件。
将文档摘要请求示例复制粘贴到你的 document-summarization.json 文件中。使用 Azure 门户存储帐户容器实例中的值替换 {your-source-container-SAS-URL} 和 {your-target-container-SAS-URL}：

请求示例

{
 "kind": "ExtractiveSummarization",
 "parameters": {
      "sentenceCount": 6
  },
 "analysisInput":{
      "documents":[
          {
        "source":{
          "location":"{your-source-blob-SAS-URL}"
        },
        "targets":
          {
            "location":"{your-target-container-SAS-URL}",
          }
          }
      ]
  }
}

运行 POST 请求

在运行 POST 请求之前，将 {your-language-resource-endpoint} 和 {your-key} 替换为你的 Azure 门户语言资源实例中的终结点值。

重要

完成后，请记住将密钥从代码中删除，并且永远不要公开发布该密钥。对于生产来说，请使用安全的方式存储和访问凭据，例如 Azure Key Vault。有关详细信息，请参阅 Azure AI 服务安全性。

PowerShell

 cmd /c curl "{your-language-resource-endpoint}/language/analyze-documents/jobs?api-version=2023-11-15-preview" -i -X POST --header "Content-Type: application/json" --header "Ocp-Apim-Subscription-Key: {your-key}" --data "@document-summarization.json"

命令提示符/终端

curl -v -X POST "{your-language-resource-endpoint}/language/analyze-documents/jobs?api-version=2023-11-15-preview" --header "Content-Type: application/json" --header "Ocp-Apim-Subscription-Key: {your-key}" --data "@document-summarization.json"

示例响应如下：

HTTP/1.1 202 Accepted
Content-Length: 0
operation-location: https://{your-language-resource-endpoint}/language/analyze-documents/jobs/f1cc29ff-9738-42ea-afa5-98d2d3cabf94?api-version=2023-11-15-preview
apim-request-id: e7d6fa0c-0efd-416a-8b1e-1cd9287f5f81
x-ms-region: China North 2
Date: Thu, 25 Jan 2024 15:12:32 GMT

POST 响应 (jobId)

你将收到 202（成功）响应，其中包含只读的 Operation-Location 标头。此标头的值包含一个 jobId，可以查询它以获取异步操作的状态，并使用 GET 请求检索结果：

显示 POST 响应中的操作位置值的屏幕截图。

获取分析结果（GET 请求）

成功 POST 请求后，轮询 POST 请求中返回的 operation-location 标头以查看已处理的数据。

下面是 GET 请求的结构：

GET {cognitive-service-endpoint}/language/analyze-documents/jobs/{jobId}?api-version=2023-11-15-preview

运行该命令之前，请进行以下更改：
- 将 {jobId} 替换为 POST 响应中的 Operation-Location 标头。
- 将 {your-language-resource-endpoint} 和 {your-key} 替换为你的 Azure 门户中语言服务实例中的值。

Get 请求

    cmd /c curl "{your-language-resource-endpoint}/language/analyze-documents/jobs/{jobId}?api-version=2023-11-15-preview" -i -X GET --header "Content-Type: application/json" --header "Ocp-Apim-Subscription-Key: {your-key}"

    curl -v -X GET "{your-language-resource-endpoint}/language/analyze-documents/jobs/{jobId}?api-version=2023-11-15-preview" --header "Content-Type: application/json" --header "Ocp-Apim-Subscription-Key: {your-key}"

检查响应

你将收到包含 JSON 输出的 200（成功）响应。 status 字段指示操作的结果。如果操作未完成，status 的值为 "running" 或 "notStarted"，你应当采用手动方式或通过脚本再次调用该 API。我们建议两次调用间隔一秒或更长时间。

示例响应

{
  "jobId": "f1cc29ff-9738-42ea-afa5-98d2d3cabf94",
  "lastUpdatedDateTime": "2024-01-24T13:17:58Z",
  "createdDateTime": "2024-01-24T13:17:47Z",
  "expirationDateTime": "2024-01-25T13:17:47Z",
  "status": "succeeded",
  "errors": [],
  "tasks": {
    "completed": 1,
    "failed": 0,
    "inProgress": 0,
    "total": 1,
    "items": [
      {
        "kind": "ExtractiveSummarizationLROResults",
        "lastUpdateDateTime": "2024-01-24T13:17:58.33934Z",
        "status": "succeeded",
        "results": {
          "documents": [
            {
              "id": "doc_0",
              "source": {
                "kind": "AzureBlob",
                "location": "https://myaccount.blob.core.chinacloudapi.cn/sample-input/input.pdf"
              },
              "targets": [
                {
                  "kind": "AzureBlob",
                  "location": "https://myaccount.blob.core.chinacloudapi.cn/sample-output/df6611a3-fe74-44f8-b8d4-58ac7491cb13/ExtractiveSummarization-0001/input.result.json"
                }
              ],
              "warnings": []
            }
          ],
          "errors": [],
          "modelVersion": "2023-02-01-preview"
        }
      }
    ]
  }
}

清理资源

如果想要清理并移除 Azure AI 服务订阅，可以删除资源或资源组。删除资源组同时也会删除与之相关联的任何其他资源。