In previous blogs, we looked at the new 3-part course for developers on developerWorks titled, Create Swift mobile apps with IBM Watson services. We dug into the first two tutorials, which use Sentiment Analysis and Visual Recognition. In this blog, we will focus the third and final tutorial on using the Text to Speech service.
Text to Speech can offer unique functionality and human-like qualities to your applications. As you can see in the video above, implementing speech to text does not have to be that difficult or time consuming. The process is accelerated in the video, but the steps in the tutorial are easy to follow. The challenge is providing the right kind of content to fuel the speech.
The Text to Speech service processes text and natural language to generate synthesized audio output, complete with appropriate cadence and intonation. It does this by providing a REST API to synthesize the speech audio. Enabling you to build more human-like voice and interactive capabilities into bots and virtual agents developed to interact with people.
This simple Text to Speech Demo shows how the service works. Using the sample app, you can customize your app using any one of 13 voice types across seven languages. Providing your own plain text, you can script interactions, support increased accessibility and much more.
Less bot more human
While bots and virtual agents have become popular and more common, the key is creating a human-like interaction. This is where the tutorials build on one another, enabling you to create bots that can not only speak, but understand context. And using sentiment analysis along with other cognitive services, you can build it to make assumptions based on user profiles and additional data to create truly personalized user experiences.
For example, recently at the Consumer Electronic Shows, IBM demonstrated how you can integrate virtual agents, powered by Watson services, with IoT enabled devices. In one demonstration, we created a virtual assistant for people in hotel rooms, providing a concierge-like experience: answering questions; recommending restaurants; etc. In another scenario, we created a virtual assistant for a conference room.
Another application of Text to Speech centers around supporting accessibility for people who are visually impaired. This allows you to give voice to content which might otherwise be difficult to access. Plus, it ensures accessibility on any web or mobile device.
Hopefully, now you are ready to build your own Swift-based, cognitive mobile apps. The potential applications for this technology are truly unlimited. As you learn and build, please share your ideas and applications with us so that we can help inspire others.