AI: Voice and Vision

AI is in everything these days. It’s in your phone, your car and maybe even your doorbell. So we organised a meetup to provide a better understanding of AI and its potential.

During our meetup AI Infusion (January 2020), four speakers lined up to introduce AI & machine learning (ML) and show how Voice AI and Vision AI work and can be used in apps today. Here are there talks:

When we start creating AI infused user experiences it’s good to have a basic understanding of AI first. Movies like “Her” suggest a general AI showing (super) human like behaviour. In reality the AI we know today is Narrow AI. At it’s core are trained models giving a prediction or decision based on the input you provide. You provide a photo and the model tells you if it “thinks” it’s a cat or a dog. In his talk Ruurd (dreamonward.com) shows how these models are trained using linear and logarithmic regression, what makes up a neural network and what they can do:

Voice AI is entering our lives in the shape of Siri, Alexa, Google Home and other voice assistants. Having a computer listen and talk back to you involves a series of steps. Your voice is captured by a mic, turned into an audio buffer, checked for wake-words (like Hey Siri or OK Google), turned into text and converted into structured data using Natural Language Understanding (NLU). From there you can add your own business logic and return something to your user using text-to-speech. In their talk, Kamil (@kamilafsar) and Şahan show you how these steps work and how they’ve used Voice AI in their work at Q42:

Vision AI enables us to build exciting new products leveraging image recognition. In his talk Jasper (@jbmoelker) shows the state of AI in the browser and how De Voorhoede makes rich web apps using AI services like Google Vision and AutoML:

We plan to do more AI projects this year, so stay tuned.

Want to attend our next meetup in person? Keep an eye on our Front-end Forward meetup group and check out our own event page!