Following up on the answer from Cristian Ene, remember also that 3J varies based on two main things: hybridisation (in this case aldehyde carbonyl is sp2) and dihedral angle. Both contribute to a smaller 3J coupling constant.
I think that both Cristian Eneand and Giselle Tamayo gave satisfactory answers for your question, but what Elena Yu Ladilina wrote is not even close to the fact
What Elena wrote is textbook in the sense, that thats where You start given that these groups usually - but not always - "displays" fast exchange on the NMR time scale. However, any activated proton will usually exchange with a deuterated protic solvent, We have seen a lot of these effect,. In the more subtle end we lost the signal from an acetamide moiety in a tetracycline to slow exchange with OD in CD3OD. Got it back by dissolving in CD3OH. And other examples like Ar-H in some phenolics (e.g. some flavonoids).
The point of this is, That what Rajnikant observes is probably explained by what Giselle and Christian writes in more than 99 % of the cases, there is a slight risk that the proton at the other end of the 3J-coupling has exchanged with a deuterated protic solvent.
So: Lack of coupling does not mean "lack of neighbourhood", but can mean that the neighbourhood does not live up to Your expectation. And that negative "proof" is a lot weaker than positive proof.
So i definitely vote for a combination of Giselle and Christians explanations.
The coupling that leads to splitting depends mainly on how the spin informations between the coupled protons can be transferred. These informations are transformed through the side way overlap between electrons of the orbitals of the C-H bonds. This is greatly affected by the bond angle that depends on the type of hybridization as mentioned by Giselle Tamayo.
Depending on the size of the J value, splitting may or may not be noticeable. A common examples of neighboring protons that often do not produce observable splitting, are alcohol (hydroxyl group) and aldehydic protons, which generally produce signals near 10 ppm, will often couple only weakly with their neighbors (i.e., a very small J value). The J value is too small, and then the signal near 10 ppm will be appear to be a singlet, despite the presence of neighboring protons.