Exploring Azure Text-to-Speech in a Quick Demo (Full Transcript)

Speaker 1: So in this quick demo, we'll see Azure Text-to-Speech in action. What we'll look at is a console application that lets us supply a voice region and gender and we will be able to supply some text that will then be spoken through the laptop. So here, we're looking at the entry point for the console application. We can see at these lines here that we collect some user input and we input the voice region. We've got a couple of choices that are then converted to an enum value. We do the same for the gender and then we then take some user input a little bit further down here in between lines 26 and 27. We take all those inputs and we push them into a method here that we can see in line 30. We can take a quick look and at this method and go to the definition here. We've just got some simple logic that extracts out the various accents and locales based on the input or the parameters are set. Those are then used to set the voice name, which is then passed on into the text-to-speech service. So we run this application and we will be prompted for a region. We'll just make that a little bit bigger. So let's see, American, female. We'll say something like. Hello. And then we can do the same for English, then male. And just remove that typo. So that was just a very quick video of what's possible with the text-to-speech service. Get more information back on the blog. There are different types of accent and locale that you can use as well as gender. There are other speech options, such as the regular voice that we've just looked at. You also have the option of using a custom voice and what that lets you do is basically supply your own training audio to get more coverage in terms of some of the accents. Maybe Scotland would be a good example of that. So I hope you found this valuable. Let me know in the comments or drop a message on my blog or hit me up on Twitter. Thank you.