[00:00:09.

450] - Speaker 1
Welcome to Azure AI Essentials. In this video, I'll walk you through what artificial intelligence or AI can
enable and how you can harness the power of Azure AI services to create breakthrough experiences in
your applications. So let's start with the basics. What is AI? Simply put, AI is the ability of computers to
perform tasks that are typically characteristic of humans. Ai enables computers to learn from massive
amounts of dataand to interact and respond more naturally with the world. With AI, computers can now
see with computer vision capabilities that allow them to interpret the world visually through video and
images. They can listen and translate, transcribe, and interpret written or spoken language, and are even
able to carry on a dialog through chat and speech, which is known as conversational AI. And they can
reason by understanding relationships between people, things, places, and events, which enables them
to automatically detect a new usual activity in a system, personalize recommendations for an individual,
and much more. Ai was once limited to researchers and institutions, but with cloud AI services and tools,
now every developer can leverage AI to create innovative applicationsthat solve complex problems and
enhance customer experience.

[00:01:37.930] - Speaker 1
We aim to empower you to use AI in your applicationsthroughoutless of skill level, and to enable you to
develop on your terms using the tools and languages of your choice. Azure AI services are the result of
years of research. Our Microsoft research team has made significant breakthroughs in the AI categories
of vision, speech, and language. In fact, we were the first to reach parity with humans for tasks like object
recognition, speech recognition, machine translation, conversational Q&A, and most recently, image
captioning. These breakthroughs are not just academic. Customers are taking advantage of them today.
These are industry leading AI models that have been tested at scale and that provide enterprise grade
security, availability, compliance, and manageability for your most mission critical applications. In fact,
Microsoft infuses AI into many of our products, from facial recognition via Windows Hello, to real time
translation while you're speaking in PowerPoint, to advanced threat detection across multiple products.
Azure AI has helped create a billion PowerPoint designer slides, deliver 6 million personalized
experiences per day on Xbox, and transcribe 10 million hours of speech every month with Microsoft
Teams. We also have an uncompromising commitment to responsible AI. We don't use customer data to
train our models, and we ensure your data privacy is always respected.

[00:03:19.650] - Speaker 1
At every stage of our AI research, development, and deployment, we keep principles like inclusiveness,
fairness, transparency, and accountability in mind. And Azure has the best privacy controls and
responsible AI capabilities, as well as the most compliant certifications of any cloud in the world. We also
offer tools and resources to help you understand, protect, and control your AI at every stage of innovation,
including responsible ML toolkits, responsible bot development guidelines, and more. Azure's AI portfolio
has options for every developer, whether you're looking for prebuilt AI models, advanced machine
learning capabilities, or a low code no code development experience. In the next few minutes, I'll walk you
through our AI services and capabilities and how you can use them to enhance your applications with AI.
So let's start by looking at Azure Cognitive Services. Azure Cognitive Services provide the most
comprehensive portfolio of customizable AI models on the market and the first AI services to achieve
human parity in computer vision, speech, and language. Azure Cognitive Services can be grouped into
four core categories vision, language, speech, and decision. All it takes is an API call to embed the ability
to see, hear, speak, search, understand, and accelerate decision making into your apps.

[00:04:57.600] - Speaker 1
And you can also customize AI models using your own data without any machine learning expertise
required. These models can also be deployed to containers, so you can run them anywhere. Let's start
with a Vision. Vision AI models enable your apps to identify and analyze content within images and
videos. Embed prebuilt models into your apps to help label content, extract printed and handwritten text,
and recognize objects. You can also create a custom vision model for your specific domain and scenario
simply by uploading and labeling a few images. Ge Aviation uses Azure computer Vision to quickly
convert handwritten and printed documents to digital format, providing a single source of truth for their
maintenance record system. Next, we have language models to build applicationsthat can understand
natural language. Create custom language models to enable users to interact with your applications, bots,
and IoT devices by using natural language. You can also translate text in real time across over 70
languages to support translation for call centers, multi lingual conversational agents, and more. And take
advantage of prebuilt models which can identify key phrases and entities in text, and perform sentiment
analysis so you can better understand customer perception.

[00:06:29.640] - Speaker 1
Developers at the BBC used Azure's language understanding service to help their virtual assistant
understand user requests, and we're able to train a custom model with relatively few examples. The
speech models allow you to integrate speech processing into your apps and services. Save time by easily
transcribing speech to text, whether you're cataloging customer service calls or creating a voice enabled
virtual assistant. Translate speech in real time and build custom models to translate terms specific to your
business. Leverage over 110 voices to build applications that can speak naturally or customize a voice
unique to your brand. Using Azure speech to text and text to speech models, Motorola Solutions helps
emergency first responders gain faster access to important information through a voice powered virtual
assistant. Finally, decision models enable you to make smarter decisions faster, create safe and positive
user experiences with content moderation capabilities that can detect and filter potentially offensive or
inappropriate images, text, and videos. Deliver personalized experiences for every user of your website or
application with models that improve over time based on user behavior, and build apps that can help you
identify potential problems before they happen. Airbus is running Azure's decision AI models in containers
to help identify potential issues with their satellites, accelerating the troubleshooting process.

[00:08:13.110] - Speaker 1
These cognitive services are powerful building blocks for developers to add AI to applications. For
business users, we provide access to these same AI models through AI Builder, which provides a no
code experience to train models and integrate them into apps within Microsoft's power platform. For
common solutions like chat bots and AI powered search, we've gone a step further, providing services
which accelerate development for these solutions. These scenario specific services often bring together
multiple cognitive services along with business logic and a user interface to solve for a common use case.
Let's take a look at just a few of the scenario specific services Azure provides. Azure Bot Service enables
you to develop intelligent bots that help you enrich your customer experience. Build a Q&A bot in just a
few clicks, or create your own branded virtual assistant using the Open and Extensible bot framework. To
better engage their customers, insurance provider Progressive used Azure Bot Service to build a chat bot
with the voice and personality of their iconic spokesperson, Flo. Next, we have Azure Cognitive Search.
Today's organizations have a wealth of information at their fingertips, but it can be difficult to search and
make sense of all this content.