What are cognitive services?
Azure Cognitive Services lets developers create applications that can see, hear, speak and understand.
They are the APIs, SDKs, and services available to help developers build intelligent apps without “direct artificial intelligence” and data science skills/knowledge.
Azure Cognitive Services let developers easily add cognitive features to their apps.
Azure Cognitive Services lets developers create applications that can see, hear, speak, understand, and even begin to reason. The Azure Cognitive Services catalog can be divided into five key features: Vision, Voice, Language, Web Search, and Decision.
Set of services that recognize the incorporation of vision capabilities in apps. The services use visual data processing to tag content (from objects to concepts), extract print and handwritten text, recognize elements known as marks and landmarks, and moderate content.
The service uses visual data processing to tag content (from objects to concepts), extract printed and handwritten text, recognize elements known as marks and landmarks, and moderate content.
Facial recognition service whose characteristics include: face detections, attributes in an image; person identification, perceived emotions recognition, and grouping of similar faces recognition in images.
Precisely extract text, key-value pairs, and tables from documents, you have the facility to adapt your recognition capabilities to documents, both on-premises and in the cloud. Let's convert forms into data.
Automatic extraction of metadata from video and audio files. It can extract spoken words, written text, faces, speakers, celebrities, emoticons, themes, marks, and scenes.
Service that recognizes digital pen content such as handwriting, shapes, and the layout of documents with digital ink inputs.
Image recognition customization to adapt to the needs of each project or service. Training the machine vision model by simply loading and tagging some images.
This service suite converts speech to text and text to speech that sounds natural. Translation between languages and the possibility of facilitating recognition and verification of the speaking user.
Speech to text
Service is used to convert audio from a wide variety of sources into text in a fluid way. It lets customization of the models to remove common barriers to speech recognition, such as specific vocabulary, speaking styles, or background noise.
Text to Speech
Allows applications and services to express themselves in a natural way, offering a wide variety of voices in a wide range of languages. Real voices with Neural Text-to-Speech functionality built into advanced research in speech synthesis technology.
Voice service easily integrates a real-time voice translation into your applications. With the possibility of personalization incorporating own translations.
API whose use is to determine the identity of an unknown speaker. The incoming audio from the unknown speaker is paired against a group of selected speakers, and if a match is found, the identity of the speaker is returned.
Set of APIs that allowed applications to process natural language with pre-compiled scripts, evaluate sentiments, and learn to recognize what users want.
Service that uses the text reading and comprehension features in applications, providing reading capacity with features such as reading aloud, translating into other languages, and attracting attention through highlighting and other design elements.
Language Understanding Intelligent Service (LUIS)
Based on Machine Learning, its objective is to provide the app with a natural linguistic understanding of applications, bots, and IoT devices.
API whose objective is to create a question and answer conversation layer with data that you already have, creating a knowledge base extracting questions and answers from semi-structured content, such as frequently asked questions, manuals, and documents. The knowledge base is getting smarter because it continually learns from user behavior.
When passing a test detects the language in which it is written, its sentiment, key phrases, and name translator Text
Neural machine translation service. It can be easily integrated into applications, websites, tools, or any solution that needs support in various languages, as well as website localization.
Support services to make smarter decisions in the shortest times by analyzing the data, obtaining statistics, and detecting anomalies in them.
Anomaly detection capabilities in apps, in order to quickly identify problems. Through an API, the preview version of Anomaly Detector receives time series data of all types and selects the anomaly detection model.
Machine-assisted content moderation API and human review tool for images, text, and videos. Detection of potentially offensive or unwanted images, filtering of unwanted text, moderation of adult content in videos, with review tools to improve results.
Artificial intelligence service that offers a personalized experience for each user, giving priority to content, designs, and conversations relevant to the user.
Possibility of including Bing search services in developed apps.