Azure Cognitive Services for AI

What is Azure Cognitive Services?

Without having specific AI or data science skills or knowledge, developers can incorporate cognitive intelligence into their applications using Azure Cognitive Services, a cloud-based artificial intelligence (AI) platform. They are accessible via client library SDKs and REST APIs in well-known development languages. With cognitive solutions that can see, hear, speak, and analyze, Azure Cognitive Services makes it simple for developers to incorporate cognitive features into their applications.

Categories of #Azure Cognitive Services: Cognitive Services can be categorized into five main pillars:

1.  Speech

·      Speech to text: You can convert audio files into text asynchronously or in real time by using speech-to-text. From a variety of sources, such as microphones, audio files, and blob storage, convert audio to text. Analyze the speaker's diary to find out who said what and when. Transcripts with automatic formatting and punctuation are readable.

·      Text to speech: You can turn input text into human-sounding synthesized speech using text to speech technology. Use neural voices, which are powered by deep neural networks and have human-like voices. Adjust the pitch, pronunciation, speaking rate, volume, and other aspects of the voice using the Speech Synthesis Markup Language (SSML).

·      Speech translation: Speech translation enables real-time, multilingual translation of speech to your applications, tools, and devices. Use this function to translate speech-to-speech and speech-to-text.

·      Speaker recognition: Speaker recognition provides algorithms that enable verification and identification of speakers based on their distinctive voice traits.

2.  Language

·      Name entity recognition: One of the features provided by Azure Cognitive Service for Language, a set of cloud-based machine learning and #AI algorithms for creating intelligent applications involving written language, is named entity recognition (NER). Unstructured text can contain entities that the NER feature can recognize and classify. Examples include people, places, businesses, and quantities.

·      Sentiment analysis and Opinion Mining: Azure Cognitive Service for Language, a set of cloud-based machine learning and AI algorithms for creating intelligent applications involving written language, offers features like sentiment analysis and opinion mining. These features enable you to learn what people think of your brand or topic by mining text for clues about positive or negative sentiment and correlating them with particular aspects of the text.

·      Question Answering: Answering questions offers cloud-based #Natural Language Processing (NLP), which enables you to overlay your data with a natural conversational layer. It is used to locate the most suitable response for any input from your personalized knowledge base (KB) of data.

Conversational client applications, such as chat bots, social media applications, and speech-enabled desktop programs, are frequently built using question-answering techniques. Several new features have been added, including end-to-end region support, precise answers, and improved relevance using a deep learning ranker.

 ·      Conversational language understanding: One of the specialized features offered by Azure Cognitive Service for Language is conversational language comprehension. It is a machine learning-based cloud-based API service that enables you to create natural language understanding components for use in end-to-end conversational applications.

Conversational language understanding (CLU) allows users to create unique natural language understanding models to anticipate the overall intention of an incoming utterance and extract valuable information from it. CLU doesn't take any action; it only provides the intelligence to understand the input text for the client application.

 ·      Translator: Any operating system can use the cloud-based neural machine translation service Translator Service, which is a member of the REST API family of Azure Cognitive Services. You can create intelligent, multilingual solutions for your applications using Translator in all languages that are supported.

 3.  Vision

·      Computer Vision: Computer Vision service provides you access for advanced cognitive algorithms for processing images and returning information. The following four computer vision services are offered.

a.  Optical Character Recognition (OCR): Text extraction or text recognition are other names for optical character recognition (OCR). You can extract printed or handwritten text from images, including posters, street signs, and product labels, as well as from documents, including articles, reports, forms, and invoices, using machine-learning-based OCR techniques. The scanned text is typically converted into words, text lines, paragraphs, or text blocks so that it can be accessed digitally. The requirement for manual data entry is thus eliminated or greatly diminished.

b.  Image Analysis: Your images can have a wide range of visual features extracted from them using the Computer Vision Image Analysis service. For instance, it can find particular brands or objects, identify human faces, or determine whether an image contains adult content.

c.  Face: #AI algorithms are provided by the Azure Face service and can find, identify, and examine human faces in images. In many different situations, such as identity verification, touchless access control, and face blurring for privacy, facial recognition software is crucial.

d.  Spatial Analysis: the use of computer vision spatial analysis to find people in videos and track their movements. ingest camera video streams, glean insights, and produce events that other systems can use. The service can perform tasks like counting how many people enter a space or checking whether face mask and social distancing rules are being followed. You can discover how people use physical spaces and increase the value of the space to your organization by processing video streams from those spaces.

·      Custom Vision: #Azure With the help of the image recognition service Custom Vision, you can create, use, and enhance your own image identifier models. Images are given labels by an image identifier based on their visual traits. Each label designates a category or item. Custom Vision, in contrast to Computer Vision, lets you define your own labels and train unique models to recognize them.

·      Face: The Face service gives users access to advanced face algorithms that facilitate the identification and detection of face attributes.

4.  Decision

·      Anomaly Detector: With the help of a set of AI APIs called Anomaly Detector, you can monitor and identify anomalies in your time series data with little to no machine learning (ML) expertise using batch validation or real-time inference.

 ·      Content Moderator: An AI service called Azure Content Moderator enables you to manage potentially offensive, dangerous, or undesirable content. It includes the artificial intelligence (AI)-powered content moderation service, which scans text, images, and videos and automatically applies content flags.

 ·      Personalizer: With the help of reinforcement learning, Azure Personalizer, an AI service, can help your applications make better decisions at scale. In order to make the best choice, Personalizer analyzes information about the state of your application, the scenario, and/or the users (contexts), as well as a set of potential decisions and associated attributes (actions). Personalizer receives feedback from your application (rewards) in order to enhance its decision-making speed.

 5.  Azure OpenAI Services

·      OpenAI Service: The GPT-3, Codex, and Embeddings model series of OpenAI's robust language models are accessible via REST API through the Azure OpenAI service. These models are easily adaptable to your particular task, including but not limited to content generation, summarization, semantic search, and translation from natural language to code.