AI Ladder: A Framework for Deploying AI in your Enterprise, Specialized areas of AI. There are a number of specialized capabilities in artificial intelligence today. Most of which depend very heavily on these foundational aspects of machine learning, namely prediction and explanation. Which of course ultimately depend on algorithms that are fed by data. Some of these capabilities include visual recognition, speech recognition, text processing, tone and empathy, and robotics. Let's take a look at some of these specialized capabilities. Visual recognition. Any AI system that processes visual information such as movies or pictures, usually relies on visual recognition algorithms to do that job. These algorithms are capable of identifying specific objects in a video or picture, or simply categorizing images based on their content. Speech recognition is the ability of a machine or program to identify words and phrases in spoken human language and then convert them into a machine-readable format. Natural language processing is the science that's used in speech recognition algorithms to recognize what words are being said. Text processing algorithms can understand complex concepts and identify entities, patterns and trends across millions of articles and other documents to provide valuable information. Again, natural language processing is used in text processing algorithms to enable these machines to read and, at a certain level, understand the content of millions of unstructured text documents. Tone and empathy algorithms are related to natural language processing algorithms. They aim to read and respond to emotions that are expressed in text or in pictures. And to improve communications between AIs and humans. And to build better AI solutions incorporating emotions as an input. Robotics deals with the design, construction, operation and use of robots, as well as, computer systems for their control, sensory feedback and information processing. Of course robotics is really big in the manufacturing sector now. But it's getting more and more complicated, as the machines are getting more sophisticated and the AI behind them is getting more sophisticated.