Stall speed is whatever airspeed you have when the wing reaches the stall angle-of-attack. When we say "stall speed", we often mean the 1-G stall speed, but under the right conditions we can place the wing at the stall angle-of-attack at a higher or lower airspeed.
The simulator may have a flaw. Also, you may not have a good understanding of what angle-of-attack is, and how it relates to pitch attitude. What exactly makes you think that the wing is not near the stall angle-of-attack whenever the stall buzzer is on? Just because the nose is below the horizon? That would be a misconception. In a high-drag, low-thrust descent, the flight path can be so steep that the nose can still be below the horizon even when the angle-of-attack is quite high.
In straight-line (unaccelerated) flight, angle-of-attack is much more closely related to airspeed than to climb rate or sink rate. In wings-level straight-line flight, for shallow to moderate climb or descent angles, the relationship between angle-of-attack and airspeed has essentially nothing to do with whether the aircraft is climbing, descending, or flying level. In these parts of the flight envelope, to a first approximation, any given airspeed is associated with with one particular angle-of-attack, regardless of the climb rate, sink rate, or pitch attitude. It sounds like this is what you are seeing on the simulator. This is normal!
Your question suggests that you may think that a descending flight path is evidence that the wing is producing less lift than in level flight, and therefore should be harder to stall, or should stall at a lower airspeed. Actually, inadequate lift is not the fundamental cause of a descent, and for shallow to moderate descent angles, the wing is only creating very slightly less lift than it would in level flight, creating an imperceptible change in the unaccelerated stall speed.
To check the functioning of the simulator, try pitching up into a very steep climb that forces the airspeed to decrease. When the stall buzzer comes on, can you make it go off immediately with a very aggressive forward push on the stick or yoke? If not, something is probably wrong with the simulator.
If everything is working right, with some practise, you should be able to fly a semi-ballistic arc with the G-load near zero as you go over the top, and with the airspeed dropping well below the normal stall speed as you go over the top, without the stall warning buzzer coming on. This will be easier to investigate if the simulator includes a G-meter.
If some parts of this answer don't make sense to you and you want to better understand the balance of forces in level, climbing, and descending flight, some of the links given at the end of this answer may help: How does an aircraft descend without its nose pointing down?
For more, see the "See How It Flies" website, especially chapter 2.