When an analogue signal is converted to digital, each sample’s amplitude must be rounded to the nearest available digital value. The difference between the actual and rounded values is the quantisation error.
At higher bit depths, this error becomes smaller and less audible. However, at lower bit depths, it can introduce noticeable distortion or graininess in quiet sounds. Dithering is used to mask quantisation error, making the resulting noise more uniform and less intrusive.