Speaker 1: It's a service which can in real time do transcription and translation and I'll show you what that looks like. You start the transcription, it takes a little while to get warmed up, so it's opening the web socket and then starting the microphone. So now, that took maybe a few seconds, so now we are recording and at the same time we're sending this off to Speechmatics and it is not only just transcribing in real time in the English here, it is transcribing in five other different languages and you can specify those languages by using the language code and you can click here to see what languages are supported but if I didn't want Italian, I wanted Korean, I'd put KO and there can be up to five of these. In this particular example, it's being set up so that after 20 seconds in the workflows, by looking at the length seconds of the recording, it will stop the recording after 20 seconds. So I just did that in this example because that's what I want to do. And then, if you look on a play button, on starting recording, there is quite a lot of options here. High console logging is really handy, it's good for debugging, so you could set this to be the way I would do it in production, I'd say if it isn't the live version, high console logging is yes and then it's no. So in production, it will be off and in development, it will be on. And a few features of Speechmatics, it can auto detect language type but you can also give it a guide as to what language you're speaking and it can do diversation and the documentation and the help site explains what all these mean. And then these are the slots, it supports up to five, so you can have five different language translations and in brief, there's fast and there's accurate, but my testing of it is fast isn't particularly fast and fast seems to be pretty accurate and accurate seems to be pretty fast and pretty accurate, so I'm not sure these make much of a difference. And there's lots of statuses that will tell you about it, vitally you need to make a backend workflow and you need to work through this and look at the example. We need this backend workflow because this is how we can, under the covers what is happening is the plugin uses a temporary token which expires, so what that means, for security reasons it's if that token expires and you never reveal your Speechmatics token on the browser. And there's some information about troubleshooting and console logging and troubleshooting client side problems, so that is about the Speechmatic plugin.