Speaker 1: Welcome to another Digital Anarchy tutorial. I'm Tor Olson, Software QA over here at Digital Anarchy, and in today's tutorial we're going to be going over how to initially set up your speech engines, specifically that of Speechmatics. So Speechmatics is one of the two speech engines that we utilize. One of them is IBM's Watson, the other one is Speechmatics. And when I am first getting started, and I pull up the Speech Engine Setup dialog box, if I could just actually hit the tilde key so I can maximize this screen, you'll see that we can enter in our credentials for our login information into Speechmatics. Now, of course, we're getting a little bit ahead of ourselves. We don't have a user ID yet, nor an API authentication token, which we need to do, but we can generate some by creating our own Speechmatics account. So that's what we're going to do first. So the first step is I would go over to my browser, where I conveniently am already on the register page. If I actually just take a step back, the first thing you're going to want to do is just initially go to speechmatics.com. That's matics with an S at the end. And from here, I'm just going to initially set up a free trial. From here, I'm going to just type in whatever email address I'm going to be using. In this case, I'm going to be setting up a co-worker of mine, so I would just type in his email address, which is Evan at Digital Anarchy. Next, I'm going to choose a password for him in order to log in. And go ahead and confirm that I'm not a robot with the CAPTCHA, and go ahead and register. All right, so now that we've made an account, we need to, of course, verify it. I also conveniently have the inbox for my co-worker's email set up in the background. And voila, I have a verification email from Speechmatics saying that I'm welcome with open arms, at which point I can confirm the email address that I've set up the account with. So when you first set up your Speechmatics account, you're going to have 360 credits, or in other words, 60 minutes to be able to play with for your transcriptions, just so you're able to take the speech engine for a spin and see what it can do. So the way that Speechmatics operates for its pricing model is that every minute is worth seven cents. But if I wanted to add more credits to be able to play with, I could just go to buy more credits, or just directly to my account tab. And from here, you can see I can add the amount of credits I want to buy, as well as add my credit card information to add those credits as well. After I add those credits, this number up here will be recalculated, allowing me to have more credits in my bank to pull from when I ask for new transcriptions. Also, while I'm on this accounts page, I can also see my user ID and my API authentication token. So these are the two bits of information we need in order to log into Speechmatics back in the transcript panel in Premiere. So the first bit of information I need is my user ID. So I'm just going to tab over and copy that, paste it in, and do the same with my API authentication token. I'm just going to keep it remembered if I want to. If I ever want to not have my username and password remembered, I can just uncheck that box if need be. For now, I'll just check it since I know I'm going to be back soon, and go ahead and click on Confirm. The checkboxes will indicate whether or not you have successfully logged in. Go ahead and close this window. And at this point, if I just create a sequence with the clip I have here, so we've got Kevin up here. So now that I have an active sequence in my program monitor and in my timeline panel, I can then go to Transcribe, and I can again expand that to see the length of the sequence I have currently selected, the name of that sequence, and of course how many available credits I have. So I can see again here, I have 60 minutes that I can play with. Additionally, I can see the status of the Speechmatics server to know that it is in good shape and that there is no delay in the minutes it'll take to get the transcription returned to me. I can also choose several different languages. Speechmatics and Watson each have their own suite of languages, but these are the ones that are available for Speechmatics specifically. It's also worth noting that I have Australian English, UK English, and American English here as well. I can of course choose speaker diarization and choose the maximum amount of speakers I'm speaker diarization and choose the maximum amount of speakers I'm going to have. I can identify questions as well, which will add a question mark punctuation to your transcript, and from here I can go ahead and click Continue, which will encode my sequence. You should only need to wait about a quarter of the time it takes for our sequence to be played back in real time. So since this was a four and a half minute sequence, it might take about a minute or so, maybe a little more, for it to get returned back to us. So I might just fast forward right through that. And voila. My transcript has now been returned, with each individual word having its own specific time code. One other thing I'd add at the end here is that when comparing Watson and Speechmatics, Speechmatics appears to be much more accurate, even with noisy audio that has background noise or music, with of course the catch that it is a little more expensive than Watson at $0.07 per minute versus Watson's $0.02 per minute. That being said, using Speechmatics instead of Watson can be a real time saver, specifically because of how accurate your return transcript is going to be, and therefore the less cleanup that you have to do. But of course I should just let the plugin speak for itself. If you want to check it out for yourself, you can go to digitalanarchy.com, where we have free trials of Transcriptive and all of our other plugins. Again, I'm Tor Olson, Software QA over here at Digital Anarchy, and we'll see you in the next tutorial.