AI and Machine-learning are computing fields that have been on a steady rise for quite some time now. AWS, one of the world's most popular cloud providers, keeps on rolling out new AI/Machine-learning-powered services as well as enhancing old ones. In this article, we are going to talk about AWS's AI/Machine-learning catalogue. This means we going to talk about the various machine-learning offerings(services) provided by AWS, what they do, how you can use them to deliver more value to your customers and also provide learning resources for each of the services. If you are as eager to learn about these services as I do, let's get right to it.

Amazon CodeWhisperer — If you are reading this article and you are a software developer, I bet this is one of the services that will be of the most interest to you. CodeWhisperer is an AI coding buddy or companion. It was officially announced in June 2022 but was only recently made free to the public. So you can use it to generate code, provide recommendations and identify issues in your code, all without having a subscription plan. How cool is that? CodeWhisperer is trained on billions of lines of Amazon and publicly available code. Some of its key features include;

• It understands comments written in natural language (in English) and can generate multiple code suggestions in real-time to improve developer productivity.
• It provides support for popular programming languages and IDEs. CodeWhisperer provides AI–powered code suggestions for many programming languages, including Python, Java, JavaScript, TypeScript, C#, Go, Rust, PHP, Ruby, Kotlin, C, C++, Shell scripting, SQL, and Scala. You can use the service from multiple IDEs, including JetBrains IDEs, Visual Studio (VS) Code, AWS Cloud9, as well the AWS Lambda console.
• CodeWhisperer is optimized for use with other AWS services thereby making it more efficient for developers to use AWS services by providing code suggestions that are optimized for AWS APIs including Amazon Elastic Compute Cloud (Amazon EC2), AWS Lambda, and Amazon Simple Storage Service (Amazon S3).
• It has built-in security scans to help you detect application vulnerabilities.

If you are interested in learning how to start using Amazon CodeWhisperer, This Amazon CodeWhisperer workshop is a great starting point.

Amazon SageMaker — SageMaker is a fully managed service provided by AWS that enables developers and data scientists to build, train, and deploy machine learning models at scale. It provides an IDE for building and training machine learning models, as well as hosting for those models. It is built on the experience Amazon has amassed from their own products such as personalized shopping, customer recommendations, robotics and much more. Think of SageMaker as a personal fitness trainer for machine-learning models. Just as a fitness trainer provides personalized workout plans and keeps track of progress, SageMaker provides a platform for building, training, and deploying custom machine-learning models. It gives you the tools to improve your model's fitness and ensure it stays in shape with regular maintenance and updates. So, if you want to train your models to be the leanest and strongest they can be, consider hiring SageMaker as your trainer. SageMaker lets you gain from the power the Machine-learning universe has to offer without necessarily having extensive expertise in data science or machine learning. If you are looking forward to getting hands-on with this service, I'd recommend that you check out this SageMaker workshop.

Amazon Rekognition — It is a fully managed AI/ML service provided by AWS that makes it easy to add image and video analysis functionalities to your applications. Amazon Rekognition can automatically identify objects, people, text, scenes, and activities within images and videos, making it a powerful tool for image and video analysis. It can be used in a variety of applications, ranging from facial recognition for security and surveillance to image and video analysis for content moderation and compliance, and image and video analysis for research purposes. Rekognition is a highly scalable service that can handle millions of images and videos per day, making it suitable for use in large-scale applications. It also provides an API that allows you to integrate its features into your applications easily. You can use the API to analyze images and videos stored in Amazon S3, as well as those streamed through Amazon Kinesis Video Streams. If you are one of those people with a keen interest in visual data analysis, Amazon Rekognition is here for you. You can check out this Amazon Rekognition workshop to learn more about the service.

Amazon Lex — It is a fully-managed AI service provided by AWS that allows developers to create chatbots and conversational interfaces using voice and text. It uses the same automatic speech recognition (ASR) technology and natural language understanding (NLU) that powers Amazon Alexa, allowing developers to create chatbots, virtual assistants, and other conversational interfaces. Amazon Lex is similar to the famous ChatGPT in that it is designed to enable human-like interactions. With Amazon Lex, developers can create natural language interactions with their applications by defining intents, which are the actions or tasks that users want to perform, and slots, which are the pieces of information that the user provides to complete the intent. Developers can also use Amazon Lex to automate common customer service tasks, such as updating an account or resetting a password. It integrates with other AWS services, such as Amazon Lambda, DynamoDB, and S3, allowing developers to build powerful and scalable applications with ease. Additionally, Amazon Lex provides built-in integration with messaging platforms like Slack and Facebook Messenger, making it easy to add conversational interfaces to these platforms. As it is said, "You learn the most by doing", be sure to check out this Amazon Lex hands-on lab.

Amazon Comprehend — Amazon Comprehend is a fully-managed natural language processing (NLP) service offered by AWS that facilitates the extraction of insights and relationships from unstructured text data. With Amazon Comprehend, users can analyze large volumes of text to identify key phrases, events, entities, sentiments, language, and even syntax, without requiring any machine-learning experience. It uses a variety of techniques such as deep learning, and rule-based analysis to understand the meaning and context of the text. It supports a range of languages, including English, Spanish, French, German, Italian, Portuguese, and Japanese. By making use of Amazon Comprehend, businesses can gain insights from their data that would be difficult or impossible to achieve through manual analysis. There is a variation of Amazon Comprehend called Amazon Comprehend Medical that is purposely designed to extract meaningful information from unstructured medical text data. It can extract medical information from unstructured text data such as doctor's notes, patient health records, clinical trial reports, and more. To learn more about Comprehend, check out this Amazon Comprehend immersion day resource.

Amazon Transcribe — It is a fully managed speech recognition service provided by AWS. It uses ASR to provide highly accurate transcriptions in real-time or from pre-recorded audio or video files making it a great tool for a range of use cases, including call centre analytics, media and entertainment, and general business transcription. Transcribe also has several features that allow users to customize the transcription output, such as speaker identification, channel identification, and vocabulary customization. With Transcribe, you can automatically remove personally identifiable information (PII) from transcription results. This feature is useful for organizations that need to protect sensitive information in their transcriptions, such as names, addresses, social security numbers, or other personal identifiers. Additionally, Amazon Transcribe integrates easily with other AWS services such as Amazon S3 and Amazon Lambda, as well as third-party services and applications. Check out this resource to learn more bout Amazon Transcribe.

Amazon Polly — Amazon Polly is a cloud-based service provided by AWS that converts text into lifelike speech allowing developers to build speech-enabled applications. It can convert any written text into a natural-sounding voice with a wide variety of voices and languages available. The service uses advanced deep learning technologies to create human-like speech patterns and intonations, which can be customized to match the specific needs of the application. It makes use of pronunciation lexicons(a dictionary that contains a mapping between words and their phonetic pronunciations) and SSML(Speech Synthesis Markup Language). With pronunciation lexicons, you can easily customize the pronunciation of specific words or phrases that may not be accurately pronounced by the default TTS(text-to-speech) engine. With SSML, you can control pronunciation, volume, and the speed of speech generated by Amazon Polly, as well as add pauses, emphasis, and other prosodic features. Polly can be accessed through a simple API call or the AWS Management Console.

Conclusion

We have explored some of the AI/ML-powered AWS services and how we can make use of these services to build applications that are more performant and offer better user experience (UX) to our user base. If you are someone that intends to keep up with the times in the technology and cloud computing landscape, I strongly believe that learning how to use these AI/ML-powered services to add value to your applications, will help you achieve your goals. It will be unfair if I end this without giving credit to the Tech With Lucy youtube channel for inspiring the article. If you are new to AWS or interested in video resources that share insights into the AWS and cloud computing ecosystem, I'll recommend you subscribe to that youtube channel.