Update for acrolinx.

Author: Sherry Yang

learn-pr/wwl-data-ai/introduction-information-extraction/includes/3-vision-extraction.md

@@ -2,7 +2,7 @@ AI-powered information extraction replaces the need to manually inspect each pie
 
 Computer vision is made possible by machine learning models that are trained to recognize features based on large volumes of existing images. Machine learning models process images by transforming the images into numerical information. At its core, vision models perform calculations on the numerical information, which result in predictions of what's in the images. 
 
-![Screenshot of an envelope with the address handwritten. The address is digitized next to the image, showing an example of OCR](../media/sample-mail.jpg) 
+![Screenshot of an envelope with the address handwritten. The address is digitized next to the image, showing an example of OCR.](../media/sample-mail.jpg) 
 
 **Optical Character Recognition (OCR)** helps computers recognize that an element in an image contains text. OCR is the foundation of processing text in images and uses machine learning models that are trained to recognize individual shapes as letters, numerals, punctuation, or other elements of text. Much of the early work on implementing this kind of capability was performed by postal services to support automatic sorting of mail based on postal codes. Since then, the state-of-the-art for reading text has moved on, and we have models that detect printed or handwritten text in an image and digitize it line-by-line and word-by-word.
 

learn-pr/wwl-data-ai/introduction-information-extraction/includes/4-form-extraction.md

@@ -4,9 +4,9 @@ Forms and other documents have text data with *semantic meaning*. Semantic meani
 
 Consider an organization that needs to process large numbers of receipts for expenses claims, project costs, and other accounting purposes. Using document intelligence, the company can take a scanned image of a receipt, digitize the text with OCR, and extract semantic meaning. The semantic meaning of data in forms can be described in field-value pairs. 
 
-- The **field name**, is the key, or type of data entry.   
-- The **feild description** is the definition of what the field name represents.  
-- The **value** cooresponds with the field name and is the data specific to the content.  
+- The **field name** is the key or type of data entry.   
+- The **field description** is the definition of what the field name represents.  
+- The **value** corresponds with the field name and is the data specific to the content.  
 
 For example, in an invoice, the fields recognized may include:
 

learn-pr/wwl-data-ai/introduction-information-extraction/includes/5-multimodal-extraction.md

@@ -1,6 +1,6 @@
 AI-powered information extraction techniques can be combined to perform data extraction on multiple modalities of content, from documents to video and audio. Using multimodal data extraction can help with digital asset management, workflow automation, generating further insights, and more. 
 
-The orchastration of techniques can include vision and document intelligence, and others including:
+The orchestration of extraction techniques can include vision and document intelligence, and others including:
 
 - **Natural language processing** can be used to find key phrases, entities, sentiment, etc. in written or spoken language.
 
@@ -10,11 +10,13 @@ The orchastration of techniques can include vision and document intelligence, an
 - **Speech recognition** takes the spoken word and converts it into data that can be processed - often by transcribing it into text. The spoken words can be in the form of a recorded voice in an audio file, or live audio from a microphone. 
 
 > [!NOTE] 
-> Speech recognition is covered in [Get started with speech on Azure](/training/modules/recognize-synthesize-speech)
+> Speech recognition is covered in [Get started with speech on Azure](/training/modules/recognize-synthesize-speech).
 
 - **Generative AI** can add to the data extraction process by allowing users to identify their own fields and field descriptions. It can be particularly useful when dealing with unstructured content. One example is the user-added *field* of "summary". The *value* associated with the field must be generated based on the data in the content.
 
 >[!NOTE]
-> Generative AI concepts are covered in-depth in [Introduction to generative AI on Azure](/training/modules/fundamentals-azure-ai-services/)
+> Generative AI concepts are covered in-depth in [Introduction to generative AI on Azure](/training/modules/fundamentals-azure-ai-services/).
+
+The content processing pipeline for multimodal information extraction can include layers of these extraction techniques. One example of the pipeline's output is structured insights and additional generated content.
 
 ![Screenshot of the possible components of multimodal information extraction.](../media/component-overview.png)

learn-pr/wwl-data-ai/introduction-information-extraction/includes/6-knowledge-mining.md

@@ -2,7 +2,7 @@ Knowledge mining solutions provide automated information extraction from large v
 
 In AI-powered information extraction for search, content first moves through **Document cracking**. Document cracking describes opening document formats like PDFs to extract the contents as ASCII text for analysis and indexing.
 
-The contents then move through **AI enrichment**, which implements AI on your original dcontent to extract more information. Examples of AI enrichment include adding captions to a photo and evaluating text sentiment. AI enriched content can be sent to a **knowledge store**, which persists output from an AI enrichment pipeline for independent analysis or downstream processing.  
+The contents then move through **AI enrichment**, which implements AI on your original content to extract more information. Examples of AI enrichment include adding captions to a photo and evaluating text sentiment. AI enriched content can be sent to a **knowledge store**, which persists output from an AI enrichment pipeline for independent analysis or downstream processing.  
 
 The resulting data is serialized as JSON data. The JSON populates the *search index*. The populated **search index**  can be explored through queries. When users make a search query such as "coffee", the search engine looks for that information in the search index. A search index has a structure similar to a table, known as the index *schema*. A typical search index schema contains *fields*, the field's data type (such as string), and *field attributes*. The fields store searchable text, and the field attributes allow for actions such as filtering and sorting. Below is an example of a search index schema: