GD&L, all speed indicators, as you indicate may be compensated. Older ASIs and sensing devices had inherent damping from the sensors and the mechanical display mechanism; generally these gave an adequate compromise between accuracy and usability in dynamic situations.
The newer sensors, display media, and computation allows the designer to fine tune the display or seek other compromises; in the early days of Air Data Computers and EFIS the overuse of the newfound technology led to interesting mistakes.
Modern sensor systems provide accurate source information; some smoothing can be applied prior to the ADC. On larger aircraft the pitot static systems may be cross coupled so as to average out transient errors due to local airflow / yaw during dynamic maneuvers.
Some speed tapes have relatively high damping, but any adverse effects of this can be offset by using a speed trend vector. This vector primarily replaces the loss of speed trend and acceleration cues that are inherent in dial displays via angular awareness and a much longer display scale. Trend vectors may be include a function of inertial acceleration for damping / quickening, but this requires a good balance of terms and often complicated inertial axis resolution. A poor system for example could indicate speed loss at the stall with an accelerating speed trend; this is unacceptable for a civil aircraft certification.
Some speed tapes now have stall margin indicators which are computed via an AOA input. A really clever system could use actual aircraft wt, ‘g’, configuration, and lift coefficients to correlate vane AOA with indicated speed. Less complex systems have simpler routines and make assumptions about lift characteristics based on trim position, the end results are surprisingly accurate.
Another aspect of your observation of speed fluctuation may be that larger aircraft (high mass) are less prone to speed fluctuation due to inertia; this may also be a function of the aircraft design – speed / AOA stability. In really turbulent conditions where there are large fluctuations the crew may add a small value to the required speed to counter any hazardous error; generally the crews do not worry about small fluctuations. However on smaller aircraft due to less inertia, etc, speed fluctuations can be disconcerting. The significance of the fluctuation is proportional to time, the shorter the period the less significant, the longer then the more attention that has to be paid to the parameter. Judgment of this significance has to be learnt and may be encompassed under general airmanship.