What is summarization?

This documentation contains the following article types:

• Quickstarts are getting-started instructions to guide you through making requests to the service.
• How-to guides contain instructions for using the service in more specific or customized ways.

To use this feature, you submit data for analysis and handle the API output in your application. Analysis is performed as-is, with no added customization to the model used on your data.

1. Create an Azure AI Language resource, which grants you access to the features offered by Azure AI Language. It generates a password (called a key) and an endpoint URL that you use to authenticate API requests.

2. Create a request using either the REST API or the client library for C#, Java, JavaScript, and Python. You can also send asynchronous calls with a batch request to combine API requests for multiple features into a single call.

3. Send the request containing your text data. Your key and endpoint are used for authentication.

4. Stream or store the response locally.

Text summarization uses natural language processing techniques to generate a summary for plain texts, which can be from a document, conversation, or any texts. There are two approaches of summarization this API provides:

• Extractive summarization: Produces a summary by extracting salient sentences within the source text, together the positioning information of these sentences.

  • Multiple extracted sentences: These sentences collectively convey the main idea of the input text. They're original sentences extracted from the input text content.
  • Rank score: The rank score indicates how relevant a sentence is to the main topic. Text summarization ranks extracted sentences, and you can determine whether they're returned in the order they appear, or according to their rank. For example, if you request a three-sentence summary extractive summarization returns the three highest scored sentences.
  • Positional information: The start position and length of extracted sentences.

• Abstractive summarization: Generates a summary with concise, coherent sentences or words that aren't verbatim extract sentences from the original source.

  • Summary texts: Abstractive summarization returns a summary for each contextual input range. A long input can be segmented so multiple groups of summary texts can be returned with their contextual input range.
  • Contextual input range: The range within the input that was used to generate the summary text.

As an example, consider the following paragraph of text:

"At Microsoft, we are 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 enables us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages. The goal is to have pretrained 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 achieve 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."

The text summarization API request is processed upon receipt of the request by creating a job for the API backend. If the job succeeded, the output of the API is returned. The output is available for retrieval for 24 hours. After this time, the output is purged. Due to multilingual and emoji support, the response can contain text offsets. For more information, see how to process offsets.

If we use the preceding example, the API might return these summaries:

Extractive summarization:

• "At Microsoft, we are on a quest to advance AI beyond existing techniques, by taking a more holistic, human-centric approach to learning and understanding."
• "We believe XYZ-code enables us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages."
• "The goal is to have pretrained models that can jointly learn representations to support a broad range of downstream AI tasks, much in the way humans do today."

Abstractive summarization:

• "Microsoft is taking a more holistic, human-centric approach to learning and understanding. We believe XYZ-code enables us to fulfill our long-term vision: cross-domain transfer learning, spanning modalities and languages. Over the past five years, we achieved human performance on benchmarks in conversational speech recognition."

This documentation contains the following article types:

• Quickstarts are getting-started instructions to guide you through making requests to the service.
• How-to guides contain instructions for using the service in more specific or customized ways.

Conversation summarization supports the following features:

• Recap: Summarizes a conversation into a brief paragraph.
• Issue/resolution summarization: Call center specific features that give a summary of issues and resolutions in conversations between customer-service agents and your customers.
• Chapter title summarization: Segments a conversation into chapters based on the topics discussed in the conversation, and gives suggested chapter titles of the input conversation.
• Narrative summarization: Generates detail call notes, meeting notes or chat summaries of the input conversation.

As an example, consider the following example conversation:

Agent: "Hello, you're chatting with Rene. How may I help you?"

Customer: "Hi, I tried to set up wifi connection for Smart Brew 300 espresso machine, but it didn't work."

Agent: "I'm sorry to hear that. Let's see what we can do to fix this issue. Could you push the wifi connection button, hold for 3 seconds, then let me know if the power light is slowly blinking?"

Customer: "Yes, I pushed the wifi connection button, and now the power light is slowly blinking."

Agent: "Great. Thank you! Now, please check in your Contoso Coffee app. Does it prompt to ask you to connect with the machine?"

Customer: "No. Nothing happened."

Agent: "I see. Thanks. Let's try if a factory reset can solve the issue. Could you please press and hold the center button for 5 seconds to start the factory reset."

Customer: "I've tried the factory reset and followed the above steps again, but it still didn't work."

Agent: "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."

Conversation summarization feature would simplify the text as follows:

This documentation contains the following article types:

• Quickstarts are getting-started instructions to guide you through making requests to the service.
• How-to guides contain instructions for using the service in more specific or customized ways.

Native document summarization uses natural language processing techniques to generate a summary for native documents. A native document refers to the file format used to create the original document such as Microsoft Word (docx) or a portable document file (pdf). Native document support eliminates the need for text preprocessing before using Azure AI Language resource capabilities. Currently, native document support is available for two types of summarization:

• Extractive summarization: Produces a summary by extracting salient sentences within the document, together the positioning information of those sentences.

  • Multiple extracted sentences: These sentences collectively convey the main idea of the document. They're original sentences extracted from the input document's content.
  • Rank score: The rank score indicates how relevant a sentence is to the main topic. Text summarization ranks extracted sentences, and you can determine whether they're returned in the order they appear, or according to their rank. For example, if you request a three-sentence summary extractive summarization returns the three highest scored sentences.
  • Positional information: The start position and length of extracted sentences.

• Abstractive summarization: Generates a summary with concise, coherent sentences or words that aren't verbatim extract sentences from the original document.

  • Summary texts: Abstractive summarization returns a summary for each contextual input range. A long input can be segmented so multiple groups of summary texts can be returned with their contextual input range.
  • Contextual input range: The range within the input that was used to generate the summary text.

Currently, Document Summarization supports the following native document formats:

For more information, see Summarize native documents

• Summarization takes text for analysis. For more information, see Data and service limits in the how-to guide.
• Summarization works with various written languages. For more information, see language support.

• Conversation summarization takes structured text for analysis. For more information, see data and service limits.
• Conversation summarization works with various spoken languages. For more information, see language support.

• Summarization takes text for analysis. For more information, see Data and service limits in the how-to guide.
• Summarization works with various written languages. For more information, see language support.