Which part of that is the clone brush? The top two pix, #1 & #2, were all that I could find on youtube; both had unwanted writings, but in different places.
To get the initial pictures, in Windows 10, I grabbed my screen shots by pressing [logos + shift + s] in playback mode to hide youtube progress bar. Then I saved the two images in .jpg format for maximum compatibility.
Then, what I did was to transfer clean imagery from the same area in #1 to the same area in #2 which was over-written with the writing.
The result was the lower, third image, with all unwanted writing removed.
There were a few technical details; setting a convenient brush size (about half the height of the lettering in #2); and round or square brush shape; getting close-up magnification to make sure the selected pixel in #1 and #2 were the exact same pixels (I chose the lightest pixel in the angel's eye); then backing off and reducing both images to 50% of normal size to work with, and setting hardness, density, thickness, and opacity all to 100% for one-pass results.
Then you right-click on clean copy in #1 for pick-up, and left-click on the same spot in #2 for lay-down. I 'think' that's what I did; as the vintage kid's game commercial goes; "It's all in the wrist action."
My daughter did something similar, recently, with a couple of group photos of herself and some friends. None of the photos came out great. At least 1 person in each of them had odd expressions on their face, or was looking in the wrong direction.
What she did was choose the 2 best photos, the easiest to work with, and then work in layers, erasing the face of one person, allowing their face from another photo from the series to show through from the layer beneath. No clone brush needed.
I probably would have done the same in the above images that you used, erasing the text on one layer, allowing the 2nd image to show through from the bottom layer, then merging the 2 layers, once all the text had been removed. You can also move the bottom layer around, if necessary, to ensure they line up correctly. Since those 2 images are exactly the same, except for the text, it would probably be even easier than what my daughter had to work with.