Chapter 2. Existing APIs and Libraries
Although digital voice interfaces have been around for several decades, we are really just now at the cusp of advanced voice interfaces, especially in the realm of Voice-as-a-Service (VaaS) and intelligent, near human-like voice-enabled virtual assistants. In this chapter, we will review some of the services currently available, including Amazon Alexa, Microsoft Cognitive Services, Cortana, and Google Cloud Speech API.
While there are plenty of services to choose from these days, they all offer unique functionality that you will want to consider when building out your own service or device. Some are easier than others to implement, but aren’t as flexible or customizable (yet may be sufficient for your use case). For example, Amazon Alexa might be a great choice for a virtual assistant with a lot of pre-packaged features, but if you’re looking for more straightforward speech to text, Nuance Mix or Amazon Lex may be a better option.
In addition to exploring some of the services available today, we will also touch a bit on some technical architectures, both those you might encounter as well as ones you might build while designing your own project. To start things off, let’s take a deeper look at Amazon Alexa.
With the launch of the Amazon Echo in November 2014, Amazon helped propel home-based voice interfaces into the mainstream: using its cloud-based voice assistant “Alexa,” anyone within proximity of the device can make ...