Speaker 1: now let's take a look at the speech-to-text service from IBM Watson all right so this is the one which converts audio and voice into written text for quick understanding of the content all right so I'm going to say will demo and it's going to take you to the demo page and as you can see here a little bit about the service itself. So it understands Arabic, English, Spanish, French, Brazilian, Portuguese, Japanese, Korean and Mandarin and it can convert all of these into text. Real cool. So these are the different formats supported and for this demo you can also use the microphone and you can record your audio and it'll be converted to text in real time. That's really cool. You can also play some of the samples here. So you can choose one of these models. We're gonna be English of course. So let's start the recording and you'll see what's gonna happen now. So we have started the recording and as we speak the demo is talking to the API and it's returning the text. You can see my speech being converted to text in real-time. Really really cool and if you see the accuracy it's actually pretty good. We have some more tabs here. Word timing and alternatives and this is the JSON body response which you are actually receiving. Real good. So now what we're going to do, we're going to try this from your Insomnia application. Okay so I'm going to go back here and what you need to do is click on get started for free. Make sure you're logged in into your IBM Watson account or your IBM Bluemix account and just go down make sure LID selected and say create. I have already associated the free service to my user so I'm directly going to go to my dashboard. I'm directly going to go inside my speech to text service here. So this is going to take you to the landing page of the service itself. There you go. That's the information we need. Credentials and the curl command. Okay. So simply copy the curl command here and go to Insomnia. Create a new request and simply paste the code here and it also actually automatically changes this to a POST request. Nice, it parses everything nicely and if you can see this is your information, the query, header, everything looks good. Just on this one if you find any difficulty actually pointing to the source file and things like that I would advise you to drop down and go to binary file say okay and say choose file and you can now choose any file which is audio file this one here and I'll say import and it'll say do you want to set the content type back to audio MPEG and you say yes all right that's it now just hit send and wait for the response there you go hello there to me folks hello you to me folks so that's what it thinks the audio said actually what I said was hello Udemy folks but it thinks hello you to me folks but Udemy is a different name it doesn't come from the dictionary so obviously we can understand and if If you want to try something else, go ahead and, you know, upload another file or something like that. So, let's upload this one, transcript 2, slightly bigger one, actually. And let's see what happens. Alright, it actually came back and you can see it says the text language must match the selected voice language, mixing language English text with a Spanish male voice does not produce valid results. The synthesized audio is streamed to the client. It's pretty good. was a rather large file comparatively of course and it did a pretty good job it took barely 10 seconds so really great really great so you can actually use this in your demos and your POCs and things like that and of course if you have an application front-end it's extremely easy to use this curl as we have demonstrated this here great going if you can do this you're up and running with the information of what the service does you can plan accordingly how you would like to use and how would like to learn more about this service. Great going.