Do you have an oscilloscope? If not, you could try using an audio probe to test for the presence of a noise signal. You should be able to detect noise going into the + input on the 741. If it’s there you can check downstream and see where it goes away.
It may be though that there’s not a good noise signal going in. The problem with noise circuits like this is they rely on behavior of a transistor (the one with the emitter not connected) being used in a way it’s not meant to be, and sometimes one transistor will produce noise and another one (even one nominally of the same type, maybe even coming out of the same bin) won’t.
So if you can’t find any noise going into the 741, my suggestion would be to build the noise source core
on a solderless breadboard and see if that works. If it doesn’t work with one transistor try another. You might find good and bad ones in the same bin, or you might need to buy more transistors and try those. Though really if the design’s good it shouldn’t be terribly hard to find a transistor that’ll work.
(Are you actually using the transistor type that was originally specified, or a substitute?)
You can also try varying the value of the collector resistor (the 330R one), that might get a non-working transistor to start working. Maybe just stick a 1k or 2k pot there and turn it until you get a good noise signal. Then measure the value of the resistance and put in a fixed resistor of about that value.
Once you have a noise source working on the breadboard you can transfer those components to your PCB (or stripboard or whatever you’re using) and hopefully it’ll be good.