Speaker 1: Hello, this is Aaron Bartel, and I'm going to show you how to start up and work with the application of speech-to-text that's made available by the Watson API. So the scenario here is that I'm running Node.js on IBM I, and from the browser I'm going to talk through my laptop mic and have it receive in that audio, send up the audio to Watson via an API, Watson returns back to my IBM I the text, which then in turn conveys it to the browser window. So that's done via this application, so here's my code in the IFS, here's my app.js file that corresponds to the article that I wrote for this, and then here's the username and password that I obtained from Watson. So again, this is in the IFS, it's running on my IBM I, and by the way, I'm going to be disabling this profile and password once this video is done. Haha. Take that you hackers. All right, so the first thing I need to do is go into the directory that this application is in to start it up, so let's check to see which directory I'm in. So this is a SecureShell session into my IBM I, and it's with the SecureShell Chrome browser plug-in, which I've found is quite nice. So let's change directory to speech-to-text, check the Node version quick, there we go, and then we want to do node app.js, actually let's just list what's in this directory first. So there's app.js right there, so node app.js. All right, so now it's up and running at that port, so if we open up another window, so this is now being served up by my IBM I, let's just look at developer tools here, and we can see any network connectivity that's happening back to the IBM I, and then also this is the JavaScript console, so pay attention to this because we'll see a lot of activity down here. So I'm going to scroll down, so this was again a git clone of a project from IBM that I did, and I cloned it down to the IFS on my IBM I, and now I'm going to go ahead, I've started the application up, so it's running under node on my IBM I, and I'm going to go ahead and start recording the audio. So I'm going to let my browser allow audio to come in, so notice that it is now transcribing my speech to text in the window there, so as I'm talking, each time I talk it obtains the audio and sends that to the speech to text Watson API out in the cloud from my IBM I, and then my IBM I receives that response back and then outputs it back to the browser here. Really this is a tremendous, in my opinion, a tremendous accomplishment at how fast it's doing this, and then also obviously the speech to text capability. Now this is something that we've seen on cell phones for a while, but now it's being very easily accessible by a common person for hobby purposes. So I set this up in about, probably about an hour, and I'm on the free version of the Watson API, so I've got so much API calls that I can make in a month without incurring any expense. So this is pretty stinking cool if you ask me. Oh, see, it even recorded stinking cool. I mean, as you can see, it doesn't transcribe everything exactly, but it lets us gain a perspective as to how this could be used in the future. So just imagine a scenario where you're at a doctor visit and everything could be recorded and transcribed, and eventually the transcription will become better, I'm sure, or to put it in a perspective of most businesses, what if you could record every conversation that your customer service rep has, and then that recording is stored subsequently in a DB2 table that's searchable. So now you can search every time somebody says a certain word, canceled order, or what have you. So it gives you some additional ways to procure analytics for the betterment of your business and for the betterment of making decisions on direction that your business should pursue. All right, that's it for this video. Thank you.