Read the Screen of Your Computer

Welcome to a video from Fields by Chris. That's Chris with a K. >> Welcome. I will be playing with eSpeak today, but that's not really the the point of the tutorial. eSpeak is just It's not the best-sounding text-to-voice application, but it's easy to install, it's small, and I'm just using an example, but you can use whatever text-to-speech application you want. What we're going to be doing today, I actually have a script on my computer. Let me type something. This is a test. I have it set up so that I can highlight and copy text, and then if I just hit my super key, alt, and T, it will actually test. >> This is a test. I have it set up to take that whatever's in my clipboard and read it at double speed using another voice to speech that sounds better. But I'm just using eSpeak in this video, but that's simple. We already have the text in clipboard, we just have to pipe the clipboard to whatever voice-to-text application or text-to-voice application you want. We're going to do something different today. We're going to grab a screenshot so you can grab text that's in an image maybe on your screen and have it read to you. So let's start. There's lots of applications out there that can take screenshots. We're going to look at scrot. So scrot, and then we'll give it a file. I'll say ocr.png, and then it just took a screenshot and saved it right here to this file. So let me go ahead and open that, and you can see there it is. It actually took picture of all three of my screens cuz I have three monitors by default. So if I try to run that again, it's going to tell me that file already exists. The way we can override that is just by giving it {dash} o, and it'll automatically override that file. Another thing we can do is we can actually choose a section of the screen we want to select. So I can say {dash} s. Now I run that, you can see my cursor has changed. I can now draw a box around part of the screen, and if I was to open, which open is just a shortcut on my machine to xdg-open, but use whatever image viewer you use. There you go. That's the new image right there from the little section of the screen that I selected. So we've been able to take a screenshot of a particular part of the screen we want. What we can do now is we can take that, and all my notes are going to be in the link in the description. We can run this command, and we can say, "Okay." Ampersand ampersand means that last command was successful. For some reason it didn't capture a screen uh shot, it's not going to run the next command. Uh, Tesseract is an application I've talked about before that does OCR, optical character recognition. It's going to take that image and look for text in it. We're going to say, "Run Tesseract again." We believe the last command was successful. We're going to run it on the image we just collected, and we're going to do just pipe it to standard output. We run that. And oh, right. I have to select a part of the screen, so let me just grab some. I'll just grab this part of the screen here. And it has converted that image to text. We do get this. Every time it will tell you estimated resolution of the image. We don't want that, so we're going to add to that. We're going to say, uh, "two." And then we'll pipe that into, uh, not pipe, but redirect that into dev null. Uh, what this is going to do is so any, um, warning messages or error messages are going to be dumped to that, so we're not going to see that. Again, now I can grab part of the screen, and it's going to now take the image that we just took of the screen, and pipe the screen the the text to the screen. Let's take that, wrap it in quotes and dollar signs. Now we're going to run that command, and the output's going to be a string. We're going to say, "Put that in a variable called text." So again, I'll grab part of the screen here. And now I can echo dollar sign text. We can see what it got from that image, and how it translated it into text. Of course, we could always put that into eSpeak. I have this turned up so you can hear it out of the speakers. That did not work. Invalid option. Dollar sign. Do I need to put this in quotations? Let's try this again. We're going to grab part of the screen. Give it to eSpeak. >> Download effects. >> There we go. >> 1000 not to be on flash tmp slash tutorial slash That's great. Let's go ahead and put that all into a script, which I actually have done. I have wrote this script. I called it OCR screen. Let me back cat that out so we can look at this. And I will put a link to this in the description, but again, it's just the lines we've done. We're running that command, putting the text into a text variable. I could uh check to make sure there's actually something in that variable. Like if it doesn't detect any text, if it's empty, it just exit. But it will echo out to the to the shell what the text is, then it will speak it, and then it will check to see if that image file exists. If so, it will remove it, just a little bit of cleanup there at the end. So now, I can run that OCR screen, and I can select something like just this here. Okay, let's try it with an actual image. Now, normally I would set this up for a keyboard shortcut. I don't have that set up yet, so I'm just going to sleep for 2 seconds and then run that command. I'm going to flip over to my web browser here where I just grabbed a random image. You can see my cursor has changed. I can now select uh this image. So, that's just a random image. I tried this on a bunch of images. Obviously, if your font gets really funky, it's going to have trouble reading it. Uh but even font like this, it did a great job. Uh so yeah, there'll be a link in the description to this script. Again, I'll put it up on the screen here. Let's go back cat. Oh, got an extra dash in there. There we go. Uh and yeah, that's the whole thing right there. You'll be able to download it from the link. Uh and I will actually put some notes to these It's only four lines of code. I'll put that in the description of the video as well. Uh and I'm In this example again, I'm using eSpeak. Uh there's other text voice to yeah, text to voice applications out there. eSpeak is just the quickest, easiest, smallest one to set up, but again, it sounds very robotic. Uh and this tutorial isn't on setting up the the voice-to-text text-to-voice application. So, I hope you found this useful. Check out the links in the description of this video. As always, I thank you for watching and I hope that you have a great day.