text processing pdfs

I am trying to build a word cloud from pdfs. Is there some sort of "demo" for this? Do I need to convert the pdfs to text first? I saw a video where he suggested converting to txt files and put them in a separate folder. ((92) Text Processing on Rapid Miner - YouTube)
I tried with a process (see attached xml) but I am getting gibberish for the output (see attached image). Any suggestions here? Thank you!