Based on the information in the question, a decrease in pitch of $1$-$2$ Hz is realistic. However, a careful assessment of the expected value is not possible without additional information on ambient temperature, some control on the fork temperature, and the exact composition of the aluminum tuning fork (usually, it is some aluminum-based alloy)
According to Wikipedia, the frequency $f$ of a tuning fork depends on the size, density, and Young's elastic modulus $E$ of the fork according to the following formula: $$ f=\frac{\alpha^2}{2 \pi L^2}\sqrt{\frac{E I}{\rho A}},\tag{1} $$ where $\alpha$ is a numerical constant ($\sim 1.875$),$L$ is the length of the prongs, $I$ is the second moment of area of the cross section, $\rho$ is the density of the fork's material, and $A$ is the cross-sectional area of the prongs. All thesethe geometric quantities and the density depend on temperature. However, the size and density vary by a few orders of magnitude, compared with less than Young's elastic modulus at normal temperature. Indeed, a typical value for the linear expansion coefficient of aluminum alloys is $\sim 2~10^{-5}$ K$^{-1}$.
Now, using theAvailable literature data allow us to estimate a rough figure of about $1$% for the relative variation of Young's modulus ($E$) in correspondence withresponse to a temperature variationchange of $10$ K ($E$ decreases whenas temperature increases), and taking into account that the frequency $f$ ofwhich dominates the tuning fork is proportional totemperature effect in the square root offormula $E$;($1$). Therefore, a decrease in frequency by $1$-$2$ Hz is a reasonable estimate for tuning forks vibrating in the range $400-500$ Hz.
Moreover, it is consistent with the fit of measurements contained in this paper if the increase of the tuning fork temperature is of the order of about $10$ K.